> Part IV — Building It: Field Operations. Everything up to this point has been the project built on paper. Estimates, schedules, subcontracts, submittal logs, logistics plans — all of it produced in an office by people wearing clean boots. Part IV...
In This Chapter
- The Hook: Twenty Minutes Before the Gate Opens
- 18.1 What "Running the Job" Actually Is
- 18.2 Layout and Survey Control in the Field
- 18.3 Material Flow: The Real Productivity Lever
- 18.4 Housekeeping: The Cheapest Productivity Program You Will Ever Run
- 18.5 Security and Loss Prevention
- 18.6 Temporary Facilities in Operation
- 18.7 Weather Management
- 18.8 The Daily Rhythm
- 18.9 Site Management for Special Conditions
- 18.10 The Interface with the Community
- 18.11 📋 Try It: Delivery Day at Northgate
- Spaced Review
- Project Checkpoint: The Willow Street Site Logistics Plan — In Operation
- Chapter Summary
- What's Next
Chapter 18 — Site Management: Layout, Logistics, Material Handling, and Running an Efficient Job Site
Part IV — Building It: Field Operations. Everything up to this point has been the project built on paper. Estimates, schedules, subcontracts, submittal logs, logistics plans — all of it produced in an office by people wearing clean boots. Part IV is where the paper meets the people. From here to Chapter 24, the unit of analysis is not a document. It is a crew of six standing on a concrete deck at 7:00 in the morning with a set of drawings, a pallet of material, and a question. Whether they are productive for the next eight hours is not determined by how good your schedule is. It is determined by whether somebody removed the obstacles between them and the work. That is what field operations is, and this chapter is where it starts.
The Hook: Twenty Minutes Before the Gate Opens
It is 6:15 on a Thursday morning in late November of Year 1 — month nine of the Northgate Outpatient Pavilion — and Margo Deacon is walking the job in the dark.
She does this every day. The gate opens at 6:30. Between 6:15 and 6:30 the site belongs to her and to the two security lights on the west laydown yard, and what she does in those fifteen minutes decides how two hundred and change people spend the next nine hours.
She comes in through the personnel gate on Kirkbride Avenue with a flashlight and a spiral notebook. Here is what she sees, in the order she sees it, and what she does about it.
6:16 — the hoist lobby, west elevation. Somebody dropped sixty sheets of five-eighths Type X gypsum board and a bundle of twenty-foot studs square across the run to the material hoist. It came in yesterday at 4:40 p.m., after Margo left, for the level-one finish mockup room. The delivery ticket is still zip-tied to the banding. The hoist opens at 6:30 and the first four loads belong to Cardinal Mechanical's duct crew on level two. Margo writes: BELLWEATHER — move by 6:45 or it's a backcharge. Ruthie: no p.m. drops at the hoist, ever.
6:19 — level two, northeast quadrant. Ashcroft Masonry has eleven people coming in at 6:30 to lay the elevator shaft walls, and their scaffold is still standing where they finished yesterday, forty feet from where they need it. Nobody moved it at the end of shift because the crew that could have moved it left at 2:30. Eleven masons and tenders at roughly $70 an hour of loaded cost, standing still for ninety minutes while they move their own staging, is about $1,155 that nobody will ever put on a cost report. Margo writes: scaffold moves happen at end of shift, not start. Put it in the huddle.
6:22 — the pour plan and the pick plan, in her pocket. Jamal Foster is placing the east half of the level-four deck topping this morning, 180 cubic yards, pump truck on the east drive, first truck at 6:30 and a truck every twelve minutes after that. And Cordova Precast has a crane pick scheduled at 10:00 at Pad A, north elevation. The crane path from Pad A to the trailer for the panel trailer crosses the east drive. So does the pump line. So does every ready-mix truck. Two people put those two things on two different schedules and neither of them looked at the other one. Margo writes: pour ends 10:30. Precast starts 11:00. Tell Ollie now, not at 10:00.
6:26 — the south fire lane. A box truck from a plumbing supplier is parked in it. Alder Street is the fire apparatus access road for both our building and Meridian's clinic, and it is the route an ambulance takes to the clinic's canopy. There is no delivery scheduled for that address and no name on the truck's dash card. Margo photographs it, radios Ruthie, and writes: Frank Petrosyan will shut this job down over a blocked fire lane and he will be right.
6:29 — the gate. A young man is standing there in new boots with a hard hat that has no orientation sticker on it. He has been sent by a subcontractor who told him to "just show up." Margo does not send him home. She walks him to the trailer, hands him off to Dani Okonkwo for the ninety-minute site orientation, and writes down the subcontractor's name in a column of her notebook that she reviews with me every Friday.
Five items. Fifteen minutes. Not one of them is a technical problem. Every one of them is a reason somebody would have stood still today.
At 6:31 I come through the gate with two coffees and Margo says the sentence I have heard from her, in one form or another, for eleven years.
"You know what my job is? My job is to remove the reasons people are standing still. That's it. That's the whole thing. Everything else is a hobby."
That is the thesis of this chapter, and it is the most useful sentence about field operations I know.
🏃 Fast Track: If you have run a job before, skim §18.1 and §18.2, then go to §18.3 (material flow) and do the arithmetic in §18.3.3 on double handling — it is the section that pays for the chapter. Then §18.7 (weather) and the
📋 Try itdelivery-day drill in §18.11.🔬 Deep Dive: Crane selection, production rates, and own-versus-rent economics are Chapter 21. Scaffold, formwork, and shoring are Chapter 22. The systematic version of everything in this chapter — making work ready, constraint logs, and flow — is Chapter 27, and if you read only one chapter after this one, read that one. Model-based layout and layout robots are Chapter 35.
Everything in this chapter — Kestrel Construction Group, the Northgate Outpatient Pavilion, Meridian Health System, and every person and subcontractor named — is a Tier-3 illustrative composite assembled from real projects. The numbers behave like real numbers. The names are invented.
18.1 What "Running the Job" Actually Is
Ask a hundred people what a superintendent does and ninety of them will say some version of "supervises the work." That is wrong in an interesting way.
On Northgate, roughly 85% of the craft hours belong to subcontractors — companies with their own foremen, their own payroll, their own means and methods, and their own bosses. Margo does not supervise their people. She cannot; directing another employer's means and methods is legally hazardous, and Chapter 19 is built on that distinction. So if she is not supervising the work, what exactly is she producing?
Here is the answer, and I want you to hold onto it for the rest of Part IV:
A superintendent's product is a site where every crew can work productively today and can work productively tomorrow.
Not a clean site. Not a fast site. Not a site where everyone likes her. A site where the work is available — where the area is ready, the material is there, the information exists, the access is open, the preceding trade is finished, the inspection has been passed, and nobody has to go find somebody to get started.
Everything else in this chapter is a technique for producing that condition.
Where the paid hour actually goes
Now the uncomfortable part. When you buy a craft hour, you are not buying an hour of installation. You are buying an hour of a person's presence, and only part of it turns into building.
People have been doing work sampling on construction sites for decades — contractors, universities, productivity consultants, owners' auditors. The results scatter widely, because a sheet-metal crew hanging duct in an open bay and a plumber trimming out a patient toilet are not the same animal, and because everybody defines "direct work" a little differently. Anyone who quotes you a precise percentage as an industry fact is selling something.
But the studies agree on the shape, and the shape is what you can manage:
| Category | What it actually is on a Tuesday | Rough share of the paid hour (illustrative) |
|---|---|---|
| Direct work — "tool time" | Hands on the actual installation | Commonly well under two-thirds; often near half |
| Waiting | For material, for a lift, for information, for access, for another trade to finish, for an inspection, for a decision | Usually the largest single loss |
| Travel | To the work, to the gang box, to the material, to the toilet, to the break area, back again | Substantial and almost invisible |
| Material handling and searching | Unbanding, moving, restacking, hunting for the right piece in the wrong stack | Substantial, and mostly self-inflicted |
| Rework | Doing it again because it was wrong, or because somebody else made it wrong | Small percentage, large dollars |
| Instruction, breaks, and administration | Huddles, toolbox talks, layout, breaks, cleanup, timekeeping | Necessary; part of it is not waste |
Three honest observations about that table.
First, some of the "loss" is not loss. A toolbox talk is not waste. Neither is a break, or the five minutes a foreman spends explaining the sequence. You are not trying to drive non-direct time to zero; that ambition produces hurt people, which is theme 4 and the subject of Chapter 24.
Second, waiting is the big one, and it is almost entirely yours. A crew waits because somebody upstream did not finish something. That somebody is frequently the general contractor.
Third, the leverage is enormous and nobody claims it. Moving direct work from, say, 45% of the hour to 52% is a 15% improvement in output for the same payroll. Nobody in this industry gets that from working faster. You get it from removing obstacles.
The rest of this chapter is organized by that table. Layout (§18.2) attacks rework and waiting on information. Material flow (§18.3) attacks waiting, travel, and handling — the three biggest bars. Housekeeping (§18.4) attacks travel, searching, and injuries. Temporary facilities (§18.6) attack travel and waiting. The daily rhythm (§18.8) attacks all of them at once.
💰 Money check: what standing still costs on a job with 210 workers.
Northgate peaks at 210 craft workers in week 61. Use a site-wide blended fully burdened craft rate of $66.00 per man-hour — an average across everybody from a laborer at $52 to a sheet-metal journeyman at $89. Your market's number will be different; run this with yours.
| Step | Arithmetic | Result |
|---|---|---|
| Site-wide labor burn, one hour | 210 workers × $66.00/MH | $13,860 per hour |
| One hour lost by everyone, one day | Same | $13,860 |
| Northgate's daily schedule exposure | $5,150 extended general conditions + $5,500 liquidated damages | $10,650 per calendar day |
Read those last two rows again. One hour of site-wide standing still costs more than one day of being late. A blocked route, a hoist that goes down, a gate that jams, a power outage on two floors — any of those can burn an hour across a large share of the site before lunch.
Now the honest footnote, because you deserve it: Kestrel does not write a $13,860 check. Roughly 15% of those hours are Kestrel's self-perform crews — call it 30 people, or $1,980 an hour of direct, immediate, cash-out-the-door cost. The other $11,880 belongs to twenty-odd subcontractors. You pay for that money in three slower, uglier ways: a disruption claim (Chapter 33), a crew that quietly gets smaller because the foreman moved his best people to a job where they can work, and a bid on your next project that is four points higher because that subcontractor remembers you.
The check you do not write is still a bill you pay.
🔄 Check your understanding. A superintendent tells you she "keeps the crews moving." What is she actually producing, stated as a condition rather than an activity?
Answer
A site where every crew can work productively today and tomorrow: the area is ready, the material is present and in the right place, the information exists, the access is open, the preceding trade is done, the inspection has passed, and nobody has to go find someone to get started. "Keeping crews moving" is the symptom. Removing constraints before the crew arrives is the job.
18.2 Layout and Survey Control in the Field
Everything in a building is located relative to something else, and the "something else" is a small number of physical points that somebody established early and that everybody now trusts. Get that wrong and you do not get a small error. You get an error that compounds, silently, for months.
The control network
Survey control is a set of permanent, protected reference points from which everything on the site is measured. On Northgate it consists of:
- Two horizontal control monuments set outside the building footprint, outside the excavation influence, and outside any place a truck can hit them — one at the southwest corner of the site near Alder Street, one at the northeast corner. Both are rebar-and-cap in concrete with a witness post and a painted circle around them.
- A project benchmark: a fixed elevation reference tied to the site datum, set on a permanent object that will not move — on Northgate, a bolt on the existing clinic's loading-dock curb, which is not going anywhere for the duration.
- A building grid: the column lines (A, B, C… and 1, 2, 3…) from the structural drawings, established on the ground as offset lines a fixed distance outside the building so they survive excavation, formwork, and steel erection.
That word offset is the practical heart of it. You cannot put a control point on a column line, because the column will be built on top of it. So you set the line five feet, or ten feet, outside, and every measurement in the field is "grid line plus the offset." Write the offset on the stake, on the hub, and on the layout sheet. Every time. The single most common layout disaster I have seen is a crew that built to an offset line thinking it was the grid line.
Elevation control and getting it up the building
Horizontal position is the easy half. Elevation is where jobs get hurt.
On a four-story building you cannot shoot every floor from the benchmark on the ground — you lose line of sight and you lose accuracy. So you transfer elevation upward: from the site benchmark to a floor benchmark on level one, then to level two, and so on. Each transfer is a chance to add error.
Two rules that will save you:
- Transfer from the source, not from the last transfer. Carry the elevation up from the original benchmark to each floor independently where you can, using a laser plummet through a floor opening or a total station on a known point. Chaining floor to floor to floor means level four inherits every mistake made on levels one, two, and three.
- Set a permanent, protected floor benchmark on every level, mark it in paint on a column with the elevation written next to it, and publish it. Every trade shoots from the same mark or you will have a ceiling grid at one elevation and a door head at another.
📊 Diagram (described): how tolerance accumulates.
Imagine four independent operations, each within an entirely acceptable tolerance:
BENCHMARK ──▶ FLOOR BM L2 ──▶ FLOOR BM L3 ──▶ CEILING GRID ──▶ DOOR HEAD
0 +1/8" +1/8" -3/16" +1/8"
│
accumulated: ~ +1/8 + 1/8 - 3/16 + 1/8
= + 1/8" (if they cancel)
= + 9/16" (if they stack)
Each step is fine. The stack is not. Nine-sixteenths of an inch is the difference between a door that swings and a door that binds, between a ceiling grid that meets the curtain-wall head trim and one that does not.
Tolerances do not average out; they accumulate in whichever direction they happen to point. The defense is to re-establish from control at each stage rather than measuring off the last thing somebody built. In a healthcare building with hard lid ceilings, medical gas rough-ins, and imaging equipment with real dimensional requirements, this is not academic. It is the difference between a smooth ceiling installation and eight weeks of arguing about who owns a three-quarter-inch problem.
Who lays out what
This is a contract question before it is a technical one, and it belongs in the scope sheets you wrote in Chapter 16.
| Layout task | Who does it on Northgate | Why |
|---|---|---|
| Site control monuments, benchmark, building offset lines | Kestrel's surveyor (subcontracted licensed surveyor) | Legal boundary and datum; needs a license in most jurisdictions |
| Building grid on each floor deck | Kestrel field engineering (Dani) | Everyone builds from it; the GC owns the common reference |
| Column and anchor-bolt layout | Kestrel field engineering, verified by survey | Steel is unforgiving and expensive to fix |
| Interior partition layout | Bellweather Interiors (framing subcontractor) | They install it; they own its accuracy |
| Sleeves, penetrations, hangers, embeds | Each MEP trade, from the coordinated model | Coordinated in the model first; see Chapter 35 |
| Equipment pads and imaging-suite rough-ins | Kestrel field engineering with the vendor's template | Owner-furnished equipment tolerances; nobody else will own it |
⚖️ What the contract says. "Contractor shall provide layout" appears in almost every subcontract and means almost nothing until you define it. Your scope sheet should say, in this order: (a) what Kestrel provides — grid lines, offsets, and a floor benchmark on each level, at a stated frequency; (b) what the subcontractor provides — all layout of its own work from the lines Kestrel provides; (c) who verifies and when; and (d) who pays for re-layout when a trade destroys the lines. That last one is the gap. A concrete crew that grinds off the level-three grid marks so a floor covering can go down has just cost three trades a day of work, and if nobody wrote down who owns re-establishing them, you will pay for it out of general conditions and call it "the cost of doing business." It is not. It is a scope gap.
🔄 Check your understanding. Your level-3 floor benchmark was set by shooting up from the level-2 benchmark, which was set by shooting up from level 1. Each transfer was within an eighth of an inch. Why is that not good enough, and what would you have done instead?
Answer
Because errors chain. Level 3 now carries level 1's error plus level 2's error plus its own, and they stack in whatever direction they happen to point rather than cancelling. Three-eighths of an inch of accumulated elevation error shows up as a ceiling grid that does not meet a curtain-wall head trim, or a door head that does not match a frame, and by the time you find it the argument is about who owns three-eighths of an inch. Transfer from the original benchmark to each floor independently — laser plummet through a floor opening, or a total station on a known point — and set, mark, and publish a permanent floor benchmark on every level so every trade shoots from the same source.
The instruments, described by what they do
I am not going to teach you buttons. Software and hardware change; the capability is what matters.
- A total station measures angle and distance to a prism (or, on modern reflectorless units, to a surface) and reports a coordinate. It is the workhorse. One person with a robotic total station can lay out points alone, which is why layout crews shrank.
- A laser level / rotating laser establishes a level plane across a room or a floor so anybody with a receiver on a rod can find the elevation. This is how ceiling grid, door heads, and topping-slab screeds actually get set.
- A laser plummet shoots a point straight up or down, which is how the grid gets from level one to level four through a floor opening.
- A layout robot takes coordinates directly from the coordinated 3D model, drives itself around the deck, and marks points on the floor — hanger locations, sleeve centers, wall lines — at a rate no two-person crew can match. The advantage is not speed. The advantage is that the points come from the coordinated model, so the sleeve you drill is the sleeve the mechanical contractor's duct is actually going to use. That whole workflow is Chapter 35, and Grace Lindqvist will make the case for it better than I can.
🔍 Why this works. The reason model-based layout reduces rework is not accuracy — a good instrument operator with paper drawings is plenty accurate. It is consistency of source. When the sheet-metal contractor lays out hangers from the mechanical drawings, the sprinkler contractor lays out from the fire-protection drawings, and the electrician lays out from the electrical drawings, all three are accurate to three different documents that were never reconciled. When all three lay out from the same clash-detected model, they are accurate to the same document, and the conflicts were resolved in a conference room in April instead of above a ceiling in February. The mechanism is single-source-of-truth, not precision.
18.3 Material Flow: The Real Productivity Lever
If you fix one thing on your first job, fix this one.
Material handling is where the hours go, and unlike almost everything else in construction, it is entirely within your control. You cannot make the architect answer faster. You cannot make the ground be what the boring log said. But you can absolutely decide where a pallet of gypsum board gets set down, and that decision is worth more than most of the things you will spend your day on.
18.3.1 The delivery process
An unscheduled truck arriving at the wrong hour is one of the most expensive things that happens on a job site, and it happens ten times a week on jobs that do not manage it.
Here is what one does. It occupies the gate, so the next truck stages in the street. It occupies the crane or the forklift, so the pick that was scheduled slides. It gets unloaded wherever there is room, which is not where the material goes, which guarantees a second handling. It arrives without a receiving inspection because nobody was expecting it, so the damage on the third bundle down gets discovered three weeks later when it is installed and nobody can prove it came that way. And the driver, who is paid by the load, gets impatient and starts making his own decisions about where to back up.
The delivery request form is the fix, and it is one page. Every subcontractor submits one for every delivery, at least 48 hours ahead, to a single person — on Northgate, that is Ruthie, who ran the gate during mass excavation and who by month nine runs the whole delivery calendar.
| Field | Why it is on the form |
|---|---|
| Subcontractor and contact with a cell number | Somebody has to answer at 6:40 a.m. |
| Date and requested window (2-hour block) | Windows, not times — trucks are late |
| Material description and quantity | So the receiver knows what "complete" looks like |
| Truck type, length, and whether it needs to back in | A 53-foot trailer needs a turning radius you may not have |
| Unloading method — crane, hoist, forklift, hand, self-unloading | This is the field that reveals the conflict |
| Destination — laydown area, floor and zone, or direct-to-install | The single most valuable field on the form |
| Who receives and inspects it | Not "the GC." A name. |
| Special requirements | Permit load, escort, hazmat, temperature, security |
Two of those fields do all the work. Unloading method is where you discover that three subcontractors have all assumed they get the crane at 9:00. Destination is where you discover that a delivery has no destination, which means it is going to be double-handled, which means you should reject it until it does.
Receiving and inspection. Someone from Kestrel — usually a field engineer, on Northgate usually Dani — meets the truck, counts it against the packing slip, and looks at it before signing. Three rules:
- Never sign a clean delivery ticket for damaged material. Write the damage on the ticket, photograph it with the truck still there, and note the driver's name. A ticket signed clean is, in most disputes, the end of the argument.
- Count it. Shortages discovered at installation are shortages you own.
- Log backorders the day they appear. A backordered item is a future schedule problem wearing a shipping label. It goes on the constraint log (Chapter 27) that morning, not the week it is needed.
🏗️ From the field. On a job before Northgate, we took delivery of 340 hollow-metal door frames over three weeks and signed every ticket clean because the frames were bundled and shrink-wrapped and nobody wanted to unwrap them in the rain. Eleven months later the framing contractor opened the last four bundles and found forty-one frames with the hinge preps on the wrong side. The manufacturer's position was that we had signed for them. Ours was that they were wrapped. Theirs was better. That was $18,400 of new frames plus six days of finish-hardware delay on a corridor that had already been painted, and it started with a signature under a plastic tarp.
18.3.2 Just-in-time versus stockpiling
Every logistics decision is somewhere on a line between two extremes:
| Just-in-time (JIT) | Stockpiling | |
|---|---|---|
| Material arrives | Days or hours before installation, in installation sequence | Weeks ahead, in bulk |
| Storage cost | Near zero | Laydown area, protection, security |
| Handling | Ideally once — truck to work face | Twice or three times, minimum |
| Damage and loss exposure | Low | High — weather, theft, forklift damage, wrong-crew scavenging |
| Cash | Later | Earlier (you may be paying for stored material) |
| Risk if the supply chain hiccups | High — no buffer, the crew stops | Low — you have a buffer |
| Risk if the schedule slips | Low — you just move the delivery | High — material sits, gets damaged, gets in the way |
| Best when | The site is congested, the supplier is reliable and local, the schedule is stable | Long lead times, volatile prices, an unreliable supplier, plenty of laydown |
The textbook answer is JIT. The field answer is: it depends on which risk is bigger, and on a congested site the handling cost usually decides it.
Work it for Northgate's curtain wall. We have 38,500 SF of unitized aluminum-and-glass curtain wall in bundled units, and by month nine our west laydown has shrunk from 2.1 acres to about half an acre because the parking structure subgrade and the utility trenching took the rest.
| Option A — Stockpile 6 weeks of units | Option B — JIT, two truckloads per week | |
|---|---|---|
| Laydown required | ~9,000 SF of blocked, dunnaged, tarped storage | ~1,600 SF staging near Pad C |
| Handling | Truck → laydown → crane → wall (2 crane touches) | Truck → crane → wall (1 crane touch) |
| Extra crane time | 62 additional picks × 14 min = 14.5 crane-hours at $310/hr = $4,495 | — |
| Extra labor to receive, dunnage, tarp, and re-rig | 96 MH × $52 = $4,992 | — |
| Damage/rework allowance (glass in a laydown yard) | ~1.5% of 38,500 SF at $92/SF installed = $5,313 | ~0.3% = $1,063 |
| Cost of a supplier miss (2-week gap, crew of 8 idle 3 days) | $0 (you have buffer) | 8 × 24 hr × $81 = $15,552 exposure | |
| Direct cost carried | $14,800 | $1,063, plus a $15,552 risk you have to manage |
The arithmetic says JIT wins by about $13,700 — if the supplier performs. So the real decision is not "JIT or stockpile." It is: how do I buy JIT's savings without buying its risk? The answer is a partial buffer. On Northgate we held one week of units on site — enough to keep the glazing crew working through a single missed truck — and ran everything else just-in-time. That costs about $2,500 in handling and buys down the $15,552 exposure to something we could live with.
That is the pattern for almost every logistics decision. Not the extreme. The buffer sized to the failure you are actually worried about.
🧩 Productive struggle: does the hoist fit?
Stop and work this before you read on. Five minutes with a pencil.
Northgate has one material/personnel hoist on the west elevation. It runs from 6:30 a.m. to 3:30 p.m., and after you subtract the morning and afternoon personnel rushes, the safety inspection, and lunch, you have about nine productive hours a day — 540 minutes. A round-trip cycle (load, travel, unload, return) averages 8 minutes to level 2, 10 minutes to level 3, and 12 minutes to level 4.
Here is what four trades and Kestrel need to move next week, in hoist loads:
| Trade | Loads | Destination | Avg cycle |
|---|---|---|---|
| Bellweather Interiors — studs and board | 71 | L2 | 8 min |
| Cardinal Mechanical — duct, hangers, valves | 96 | L2–L3 | 9 min |
| Halcyon Electric — conduit, strut, panelboards | 58 | L3 | 10 min |
| Summit Roofing — insulation and membrane | 34 | L4/roof | 12 min |
| Kestrel — temp materials, debris down, misc. | 40 | mixed | 9 min |
Does it fit in a five-day week? If not, by how much, and what are your five options?
Work it through
Capacity: 540 min/day × 5 days = 2,700 minutes.
Demand:
| Trade | Loads × cycle | Minutes |
|---|---|---|
| Bellweather | 71 × 8 | 568 |
| Cardinal | 96 × 9 | 864 |
| Halcyon | 58 × 10 | 580 |
| Summit | 34 × 12 | 408 |
| Kestrel | 40 × 9 | 360 |
| Total | 2,780 |
Demand exceeds capacity by 80 minutes — at 100% utilization.
And 100% utilization never happens. A hoist that is loaded, dispatched, and returned with no dead time is a fantasy. Real utilization on a well-run job is 70–80%, which puts your true capacity at 1,890–2,160 minutes. So you are actually short by roughly 620 to 890 minutes — the better part of two full days of hoist capacity.
Somebody is going to stand still, and the only question is whether you choose who or let the fastest foreman at the call button choose for you.
Five options, cheapest first:
- Move the roofing loads to the crane. Summit's 34 loads at 12 minutes are the worst use of the hoist on the list — big, light, and destined for the roof, which the crane reaches anyway. Saves 408 minutes, about 15% of the whole problem, and costs roughly two hours of crane time.
- Pre-stage board on level 2 before the crush. Bellweather's 71 loads are not time-critical to the hour; move them to the last hour of the day and the first hour of the morning, when the hoist is otherwise light. Costs nothing but a conversation.
- Add a telehandler for levels 1 and 2. A rough-terrain forklift with a boom reaches level 2 through a deck opening for a lot less than a second hoist. Roughly $4,200/month plus an operator.
- Extend hoist hours. A second operator for two hours a day, five days: 10 hours × $52 = $520/week, and it works only if the trades will staff the receiving end.
- Add a second hoist. Erect, engineer, tie back, rent, dismantle: five figures, six to eight weeks of lead time, and there is nowhere obvious to put it. This is the option everybody proposes first and it is almost always last.
Notice what you just did. You took a vague complaint — "the hoist is a nightmare" — and turned it into 2,780 minutes against 2,700, then into a ranked list with prices. That is the job.
18.3.3 Double handling: the most expensive habit on a job site
Double handling is moving material a second time without installing it. It has no name in the estimate, no cost code of its own on most jobs, and no advocate. It is the purest waste in construction, and it is everywhere.
Let us price it on the one quantity you already know. Northgate has 412,000 SF of gypsum board — all faces, all layers.
Step 1 — convert to a handleable unit. Standard 4-foot by 12-foot sheets are 48 SF each.
412,000 SF ÷ 48 SF/sheet = 8,583 sheets. Call it 8,585.
Step 2 — price one extra handling. A two-laborer team with a drywall cart, breaking down a stack, moving it an average of 120 feet on a level floor, and restacking it against the wall where it will be hung, moves roughly 48 sheets per crew-hour on a site with clear routes. That is a crew-hour, so each is two man-hours.
8,585 sheets ÷ 48 sheets/crew-hour = 178.9 crew-hours 178.9 crew-hours × 2 workers = 357.8 man-hours, call it 358 MH 358 MH × $52.00/MH (composite laborer, fully burdened) = $18,616
Eighteen thousand, six hundred dollars to move drywall from one place on the floor to another place on the floor. Once. Nobody installed anything. Nobody got closer to done.
Step 3 — the part that actually hurts. That number assumes the second move happens on the same floor with a clear route. If any of it crosses floors, it eats hoist time, and hoist time is the scarcest resource on the job:
8,585 sheets ÷ 40 sheets per hoist load = 215 hoist loads 215 loads × 9 min average cycle = 1,935 minutes = 32.3 hoist-hours At 540 productive minutes per day, that is 3.6 full days of hoist capacity
You did not spend $18,616. You spent $18,616 plus three and a half days of everybody else's access to the building. Cardinal's duct crew, Halcyon's electricians, and Summit's roofers all paid for your unplanned drywall stack and none of them will ever know why the week felt slow.
⚠️ Safety alert. Double handling is not just a cost problem; it is an injury-generation machine. Every extra handling of a 105-pound sheet of five-eighths Type X board is another chance for an overexertion strain, a pinch, a dropped load, or a cart running over a foot. Material handling and lifting injuries are among the most common serious injuries in construction, and unlike a fall from height, they arrive quietly and cumulatively. A material-handling plan is a safety plan. When you decide where a pallet gets set down, you are deciding how many times a human being is going to pick up its contents.
18.3.4 Vertical transportation: the scarcest resource on the job
On any building more than two stories tall, vertical transportation capacity — not labor, not material, not money — is usually the binding constraint. Everything has to go up, most of it has to come back down as waste, and there is exactly one way up.
Northgate has three:
| Route | Capacity | Best for | Cost of an hour |
|---|---|---|---|
| The crane (one 90-ton hydraulic truck crane, three pads) | Heavy, bulky, roof-destined | Precast, curtain-wall bundles, rooftop units, steel, roofing | ~$310/hr all-in |
| The material/personnel hoist (west elevation, 6,000 lb) | 540 productive min/day | Palletized interior material, people, small equipment | ~$105/hr equivalent |
| The stairs | Unlimited and terrible | Hand tools, small parts, and the last resort | Free, and the most expensive thing you own |
That last row is not a joke. A crew that "just carries it up" is a crew you are paying at $66 to $89 an hour to be a freight elevator.
Crane time is a formal daily document, not an understanding. Here is the actual pick schedule for the Thursday in Margo's hook, published the previous afternoon and posted at the gate, in the trailer, and in the group text.
NORTHGATE — CRANE PICK SCHEDULE — Thursday, late November, Year 1 90-ton hydraulic truck crane · operator + oiler · Lift Director: M. Deacon
| Time | Pad | Load | Trade | Wt (ea) | Dur. | Notes / conflicts |
|---|---|---|---|---|---|---|
| 6:30 | A | Precast panels P-31 → P-34 | Cordova Precast | 14,800 lb | 90 min | Clinic drive: flagger + spotter, no swing over drive |
| 8:00 | A | Precast panels P-35 → P-37 | Cordova Precast | 14,800 lb | 70 min | Wind check at 7:45 — sail area |
| 9:10 | — | CRANE MOVE A → C | Kestrel | — | 90 min | Mats, outriggers, travel. Crosses east drive — pour must be done |
| 10:40 | C | Curtain-wall bundles, L3–L4 south | Crossline Glass | 6,200 lb | 60 min | Direct to floor staging, not laydown |
| 11:40 | C | Roof curbs + RTU rails | Cardinal Mechanical | 3,400 lb | 45 min | |
| 12:25 | — | Lunch — crane secured | — | — | 35 min | |
| 1:00 | C | Roof insulation + TPO rolls | Summit Roofing | 4,100 lb | 75 min | Wind check at 12:50 |
| 2:15 | C | Empty pallets, banding, debris box down | Kestrel | — | 30 min | Never skip this |
| 2:45 | — | Contingency — held open | — | — | 45 min | Reserved. Do not fill it in advance. |
Four things about that table are worth more than the table.
One: the crane move is a 90-minute pick that lifts nothing. Eight and a half hours of crane availability, and 90 minutes of it — 18% — is spent driving from one pad to another, setting mats, and deploying outriggers. That is the single most expensive line, and it is invisible unless you write it down. This is the whole lesson of Case Study 02.
Two: the contingency block is held empty on purpose. Something will go wrong. If your pick schedule is 100% committed at 5:00 p.m. the day before, then the first delay of the morning cascades into the afternoon and the last trade on the list — always the same trade, always the one with the least clout — gets nothing.
Three: the debris pick is scheduled. If taking waste down is not on the schedule, it does not happen, and then it is on the floor, and then §18.4 becomes a problem.
Four: the pour is on the same page. Jamal's 180-CY level-four topping pour runs 6:30 to about 10:30 with the pump on the east drive. The crane move crosses the east drive. Somebody looked at both and put the crane move at 9:10 with a note. In the hook, nobody had. That is the whole difference between a good Thursday and a bad one.
How conflicts actually get resolved. In order:
- Anything with a truck already rolling wins — you cannot un-dispatch concrete.
- Anything that blocks a life-safety route loses, always, immediately, no discussion.
- Direct-to-install beats laydown. If one pick puts material in its final position and the other puts it on the ground for a second handling, the first one goes first.
- A crew already on site and waiting beats a crew not yet mobilized.
- Critical-path work beats float work — and you should know which is which without looking, because you built the schedule in Chapter 14.
- When two things are genuinely equal, the one that has been bumped most recently goes first. Fairness is not sentimentality. It is how you keep foremen submitting honest delivery requests instead of padding them.
18.3.5 Horizontal transportation: what happens after it lands
Getting material into the building is half of it. The other half is where it lands.
Laydown adjacency. Material should be stored as close as physically possible to where it will be installed, and the priority order is: direct to the work face > same floor, same zone > same floor > laydown yard. Every step down that list adds a handling.
Route width. Your building has a circulation system before it has corridors, and you should design it deliberately. On Northgate we chalked a six-foot-wide primary route on every floor from the hoist landing to the far end, marked it in yellow, and treated it as a fire lane: nothing gets set down in it, ever. Six feet lets two carts pass. Four feet does not, and a route where carts cannot pass is a route where the second cart waits.
Floor loading limits — the one that can kill somebody. An elevated composite deck is designed for a distributed live load, not for forty skids of porcelain tile in the middle of a bay. A pallet of tile can run 3,000 pounds; forty of them in a 20-by-20 bay is 6,000 pounds per 100 SF, which is far beyond typical office or clinic design live loads. On Northgate, the rule Margo enforces without exception is: any concentrated storage over 2,000 pounds on an elevated deck gets a written OK from Caldwell Structural, and it gets stacked over a column line, not mid-bay. That is an RFI and two days. A collapsed deck is a fatality.
📊 Diagram (described): planned versus unplanned floor staging.
LEVEL 2 — STAGED BY AREA (planned) LEVEL 3 — SINGLE DROP (unplanned)
┌──────┬──────┬──────┬──────┐ ┌──────┬──────┬──────┬──────┐
│ Z1 │ Z2 │ Z3 │ Z4 │ │ │ │ │ │
│ ▓▓ │ ▓▓ │ ▓▓ │ ▓▓ │ │ │ │ │ │
├──────┼──────┼──────┼──────┤ ├──────┼──────┼──────┼──────┤
│ Z5 │ Z6 │ Z7 │ Z8 │ │ │ │ │ │
│ ▓▓ │ ▓▓ │ ▓▓ │ ▓▓ │ │ │ │ │ │
├──────┴──────┴──────┴──────┤ ├──────┴──────┴──────┴──────┤
│ ══════ 6' ROUTE ══════ [H]│ │ ▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓ [H]│
└───────────────────────────┘ └───────────────────────────┘
▓ = board staged in the zone ▓ = everything, at the hoist
where it gets hung. Average haul: 180 ft, twice.
Average haul: 25 ft, once.
Same building. Same material. Same crew. The right-hand floor costs about a quarter more in labor to hang, and Case Study 01 works the arithmetic sheet by sheet.
🔄 Check your understanding. A subcontractor's delivery request says "Tuesday a.m., 1 flatbed, ceiling grid, unload by forklift." What is missing, and why does its absence guarantee a cost?
Answer
Destination. There is no floor and no zone. Without a destination the material gets unloaded wherever there is room, which is not where it gets installed, which guarantees at least one extra handling — and probably a hoist trip you did not plan for. "Requested window" is also soft ("a.m." is four hours, not two), and there is no named receiver. Send it back with three fields to fill in. Ninety seconds of pushback now beats an hour of cart-pushing later.
18.4 Housekeeping: The Cheapest Productivity Program You Will Ever Run
I am not going to give you a homily about a tidy job site. I am going to give you the mechanism.
A cluttered floor costs you in four measurable ways:
- Travel. Every obstruction lengthens every trip. A crew that walks around a debris pile forty times a day walks an extra 1,200 feet.
- Searching. Material buried under other material is material that gets re-ordered. I have seen a job buy the same box of fire-rated sealant four times.
- Rework and damage. Finished work stored under a pile of debris gets damaged. Door frames get bent. Ductwork gets crushed. Drywall corners get destroyed.
- Injuries. Slips, trips, and falls on the same level are among the highest-frequency injuries in construction, and they come overwhelmingly from housekeeping: cords, hoses, banding, offcuts, standing water, and the twelve inches of clearance that used to be a walkway.
🔍 Why this works. The deeper reason a clean site is a productive site is not that cleanliness causes productivity. It is that housekeeping is a leading indicator of planning. Debris accumulates in areas where nobody decided who owned the area. A floor with no debris on it is a floor where somebody assigned the zones, sequenced the trades, planned the material, and scheduled the waste removal — and those things cause productivity. The mop is a symptom. When Margo walks a floor and the floor is filthy, she is not annoyed about the mess. She is reading a report that says nobody is managing this area, and she goes looking for the reason.
Who owns cleanup, in writing
This is a scope-sheet question, and it is one of the most reliable scope gaps in the business. Every subcontractor's proposal says "daily cleanup of our own debris to a Contractor-provided dumpster." Every subcontractor means something different by it, and none of them means "sweep."
Your subcontract scope sheet (Chapter 16) needs to say, per trade:
| Item | Must be defined as |
|---|---|
| Scope of cleanup | "All debris, packaging, banding, offcuts, and dust generated by this Subcontractor's work, removed from the work area to the Contractor's designated container daily, before the end of each shift." |
| Sweeping | Broom-clean of their own area daily. Say the word "broom." |
| Container | Who supplies it, where it is, and whether the subcontractor may use it for anything else |
| Separation | Which stream — see below |
| Escalation | "If Subcontractor fails to clean after one written notice and 24 hours, Contractor may clean and backcharge at cost plus 15%." |
| Rate | State the backcharge rate in the subcontract. Do not discover it during the argument. |
⚖️ What the contract says. A backcharge without notice is a dispute, not a recovery. Almost every subcontract requires the general contractor to give written notice and an opportunity to cure before performing the subcontractor's work and charging for it. Skip that step and you will have a $6,000 cleanup charge that the subcontractor rejects, that sits in the pay application as a disputed item for four months, and that you eventually trade away in a change-order negotiation for nothing. The notice is not bureaucracy; it is the entire value of the clause. One email, time-stamped, naming the area and the deadline, converts a squabble into an enforceable charge. Jurisdictions and standard forms vary on notice and setoff, so read yours and ask your attorney what your state does with disputed setoffs against a subcontractor's payment rights.
⚠️ Safety alert: the four hazards housekeeping actually controls.
Treat this as the operational half of Chapter 24, not as tidiness.
-
Slips, trips, and falls on the same level. These are among the highest-frequency injuries in construction, and they come almost entirely from housekeeping: extension cords and air hoses run across walkways, banding and offcuts, standing water, dust on a smooth deck, and material creeping into a route until twelve inches of clearance is left. Same-level falls rarely kill anyone. They produce the recordable injuries, the light-duty assignments, and the experience-modification-rate increase that raises your insurance cost on every job you bid for the next three years.
-
Access and egress. Every floor needs a clear, marked, unobstructed path to its means of egress at all times, and as a building rises you add temporary stairs, ladders, and — per fire-code requirements during construction — a temporary standpipe kept within a floor or two of the highest work. A stack of material in an exit path is not a housekeeping problem; it is a life-safety violation, and it is the fastest way to get shut down by the authority having jurisdiction. The blocked fire lane on Alder Street in this chapter's hook is the same rule applied outdoors.
-
Struck-by. Struck-by hazards are one of OSHA's Focus Four — the four hazard categories that account for the large majority of construction fatalities, along with falls, caught-in/between, and electrocution. In material handling, struck-by means a load swinging, a load dropping, a bundle sliding off a stack, a cart getting away on a ramp, or equipment backing up. In crane operations it means never allowing anyone under a suspended load, controlling the swing path, using tag lines, barricading the pick zone, and understanding that the operator has absolute authority to refuse a lift. Loose material left near a leading edge on an elevated deck is a falling-object hazard for everyone below it.
-
Overexertion and manual handling. The quiet one. Every unnecessary handling of a 105-pound sheet is another lift, another awkward posture, another chance for a back or shoulder injury that will not show up as a recordable on the day it happens.
The rule that ties them together: a material-handling plan is a safety plan. When you decide where a pallet is set down, you are simultaneously deciding how many times a person will pick it up, how far they will carry it, whether the route they take is clear, and whether anyone will be underneath it. Those are safety decisions wearing a logistics costume.
The composite crew
Even with perfect subcontract language, some debris has no owner. Somebody has to sweep the shared corridors, empty the site-wide dumpsters, pick up in the parking lot, and clean the areas where four trades worked and nobody will admit to the mess.
That is the composite cleanup crew — laborers carried in Kestrel's general conditions whose cost is then allocated back to the trades. On Northgate, the composite crew and containers are budgeted at $388,000 across the job. Two ways to handle the money:
- Composite/pro-rata: the crew's cost is split among the subcontractors by an agreed formula — usually in proportion to subcontract value or headcount — and stated in every subcontract before award. Clean, predictable, no arguments, and the honest subcontractors subsidize the sloppy ones.
- Backcharge by observation: the crew tracks hours by area and trade, the superintendent signs the tickets daily, and the charges go out monthly with the pay application. Fairer, more accurate, and it requires real discipline — an unsigned cleanup ticket from three weeks ago is worth nothing.
Most good jobs run both: a small pro-rata composite charge for genuinely shared areas, plus backcharges for specific, documented, noticed failures.
Waste separation
If your project is pursuing LEED (Leadership in Energy and Environmental Design) certification — Northgate is targeting LEED Silver — construction waste diversion is a credit, and credits are earned with weight tickets, not intentions. That means separated containers on site (typically metals, cardboard and clean wood, gypsum, and mixed/landfill), labeled in the languages your crews actually read, positioned where crews will use them rather than where they fit, and a hauler that will provide certified weights by stream. Get the separation wrong in month three and you cannot fix it in month eighteen. The full picture — diversion rates, documentation, and what the credit is actually worth — is Chapter 36.
🔄 Check your understanding. A subcontractor has left three days of debris in a corridor. Your subcontract allows you to clean and backcharge at cost plus 15%. Walk through the exact sequence, and say what happens if you skip a step.
Answer
Written notice naming the specific area and the deadline → 24 hours to cure → if not cured, clean it with your composite crew, tracking hours by area and trade on a ticket the superintendent signs that day → charge at the rate stated in the subcontract, with the notice, the ticket, and dated photographs attached to the pay application.
Skip the notice and you do not have a backcharge; you have a disputed setoff that will sit in the pay application for months and get traded away in a change-order negotiation for nothing. Skip the same-day signed ticket and you have a number nobody can substantiate three weeks later. Skip the rate in the subcontract and you will be negotiating the rate during the argument, which is the worst possible time. Note the jurisdictional wrinkle too: states differ on how disputed setoffs interact with a subcontractor's payment rights under prompt-payment statutes — read yours, and ask your attorney.
18.5 Security and Loss Prevention
A construction site is an unlocked warehouse full of copper, fuel, tools, and equipment, staffed by hundreds of people whose faces nobody knows, and it is empty sixteen hours a day.
Industry and insurance estimates of annual U.S. construction theft losses run into the hundreds of millions to low billions of dollars, with recovery rates commonly reported to be low. Treat the precise figure as unknowable. Treat the direction as certain.
What gets stolen, roughly in order: small power tools · copper wire and pipe · fuel · fixtures, appliances, and anything in a box with a brand on it · lumber and sheet goods · catalytic converters out of your own crews' trucks · and, on a bad night, a skid-steer.
What it actually costs. The theft is the small number. Here is a real shape of one, from a Monday morning on a job of mine:
| Item | Cost |
|---|---|
| Replacement material: 220 LF of 4/0 copper feeder, 6 spools of #12, two temp distribution panels | $9,400 |
| Labor to re-pull, re-terminate, re-test: 96 MH × $89/MH | $8,544 | |
| Temp power down on levels 2 and 3 for most of a day: 46 workers × 5 hr × $58 blended on those floors | $13,340 | |
| Re-inspection and the RFI to document the replacement | ~$1,200 |
| Total cost to the project | $32,484 |
| Estimated scrap value to the thief | ~$2,200 |
That asymmetry is the entire argument for spending money on security. You are not protecting $2,200 of copper. You are protecting the day.
The countermeasures, in rough order of dollars-per-dollar-returned:
| Measure | What it actually does |
|---|---|
| Lighting | The cheapest deterrent there is. Dark corners of the laydown yard are where material leaves. |
| A single controlled gate | You cannot control what leaves if there are four ways out. This is also why the one-gate constraint in §18.11 is a feature, not just a nuisance. |
| Badging and orientation stickers | A hard hat sticker means this person was oriented, is on a roster, and belongs to a company you have a contract with. The man at the gate in the hook had no sticker, and that is a security issue as much as a safety one. |
| Fencing with privacy screen | You cannot steal what you cannot see from the street. |
| Cameras with recorded, remotely monitored feeds | Deterrence and, more usefully, proof. A camera nobody watches is a prop; a monitored feed that produces a phone call at 11:40 p.m. is a tool. |
| Tool crib and check-out | Somebody's name on every tool. Bluetooth or RFID tags on the expensive ones. |
| Locked, alarmed conex boxes, anchored | Ten minutes of delay is often enough. |
| Fuel management | Locked caps, metered dispensing, and a reconciliation between gallons purchased and hours run. Fuel walks off in five-gallon cans. |
| After-hours and weekend protocol | The hard one. See below. |
The weekend problem. Most losses happen when nobody is there, and the tempting fix — a guard — is expensive and only moderately effective. What works better is making after-hours access an event: all equipment parked in one lit area with buckets down and batteries pulled, a written after-hours access list that the security company holds, a required call to the superintendent for any weekend work, and a locked gate with a keyed padlock whose keys are inventoried. And when a subcontractor does work on a Saturday, they sign in and out, because Saturday work is also when the scaffold in the canonical near-miss got modified overnight by a trade that did not erect it.
On Northgate, site security — cameras, monitoring, and after-hours patrols — is budgeted at $86,000, a little under 3% of general conditions. The first month we ran without it, before the fence went up, we lost about $11,000 in tools and material. It is not a close call.
18.6 Temporary Facilities in Operation
You estimated these in Chapter 13 as Division 01 line items. Now you have to operate them, and operating them badly costs more than estimating them badly.
Here is Northgate's temporary-facilities budget, carved out of the $2,900,000 general-conditions line in the GMP:
| Division 01 line | Northgate budget |
|---|---|
| Temporary power — service, distribution, monthly consumption | $174,000 |
| Temporary heat and winter protection | $164,000 |
| Temporary water and sanitation | $72,000 |
| Temporary lighting and task lighting | $46,000 |
| Fencing, gates, barricades, signage | $64,000 |
| Site security | $86,000 |
| Material/personnel hoist — engineer, erect, rent, dismantle | $248,000 |
| Composite cleanup crew and containers | $388,000 |
| Site logistics — flagging, gate coordination, sweeping, dust control | $104,000 |
| Subtotal — temporary facilities and site operations | $1,346,000 |
That is 46% of general conditions and just under 2.8% of the GMP. It is real money and it is nobody's favorite subject, which is exactly why it drifts.
Temporary power and GFCI discipline. Every cord, every box, every tool on a construction site is on temporary wiring in a wet, abused environment. OSHA's electrical requirements for construction (29 CFR 1926, Subpart K) require, for 120-volt 15- and 20-amp receptacle outlets that are not part of the permanent wiring, either ground-fault circuit-interrupter (GFCI) protection or a documented assured equipment grounding conductor program with scheduled inspection and color-coded tagging. Pick one, in writing, and enforce it. In practice, most contractors choose GFCI everywhere because it is simpler to enforce than a tagging program that requires monthly discipline from twenty companies. Also: temp panels get locked, cords get inspected and removed from service when the ground pin is missing, and cords do not run through doorways or standing water. Every one of those is a fatality prevented in advance.
Temporary heat and the enclosure milestone — where schedule becomes money. Northgate dries in on March 28, Year 2. Before that date, heating an unenclosed steel-and-deck building through a real winter costs:
| Component | Arithmetic | Daily |
|---|---|---|
| Indirect-fired diesel heaters, 6 units | 6 × 14 hr × 4 gal/hr × $4.10/gal | $1,378 | |
| Unit rental | 6 × $58/day | $348 | |
| Ducting, fire watch, and monitoring labor | 4 MH × $52 | $208 | |
| Fuel delivery and misc. | $16 | |
| Total | ≈ $1,950/day |
We budgeted 84 days of protection: 84 × $1,950 = $163,800, which is where the $164,000 line came from. We ran 77 days and spent $150,150, coming in $13,650 under — because the enclosure-by-area resequencing we did to recover the steel delay put curtain wall on the south elevation about two weeks early, and we closed the south third of levels one and two ahead of the milestone.
Read that as a sentence: every week you pull enclosure earlier is $13,650 of temporary heat you do not spend. That is theme 2 in its purest form. A schedule decision made in the trailer in August showed up as a cash number in February. Anyone who discusses enclosure sequencing as a schedule question without a dollar sign attached is only doing half the job.
Temporary lighting. OSHA sets minimum illumination levels for construction areas in foot-candles, scaled by task — a few foot-candles in general construction areas and access ways, more in shops and offices, considerably more at first-aid stations. Look up the current table for your work; do not carry a number in your head. The practical point is that the regulatory minimum is a floor, not a target. Nobody hangs and finishes drywall, terminates devices, or reads a tape at five foot-candles. If you do not provide adequate task lighting, the trades will bring their own and plug it into whatever they find, and now you have an extension-cord problem and a temp-power load problem you did not plan for.
Water, sanitation, and the arithmetic nobody does. OSHA's sanitation requirements (29 CFR 1926, Subpart D) scale the number of toilet facilities with the number of employees and require potable drinking water, single-use cups, and clearly marked non-potable outlets. The standard is a floor. The practical planning ratio most superintendents use is one portable unit per 10 workers, serviced weekly — at Northgate's peak, 21 units.
Now do the part nobody does. Where you put them is a productivity decision:
A worker on level 4 who must walk down four flights and 300 feet to reach a toilet is gone for roughly 12 minutes instead of 4. 8 extra minutes × 210 workers × 1 trip per day ÷ 60 = 28 man-hours per day. 28 MH × $66.00/MH = $1,848 per day.
Over a hundred days that is $184,800 — more than your entire temporary-facilities budget for water, sanitation, and lighting combined, spent on stairs. Get units onto the floors as soon as the hoist can carry them, put hand-wash stations with them, and stop treating this as a comfort item.
The same logic governs break and eating areas and drinking water. Provide a covered, lit, seated area with trash containers and crews will use it; provide nothing and you will find lunch debris inside a wall cavity that will smell for twenty years. Provide cool water at the work face in hot weather and you reduce both heat illness and the walk to get it; heat illness prevention is genuine safety territory and OSHA and NIOSH both publish current guidance on water, rest, and shade.
🔄 Check your understanding. Your project manager wants to cut the temporary-facilities budget by $60,000 to cover an overrun elsewhere. Name the two line items you would defend hardest, and give the dollar argument for each.
Answer
Temporary heat, because it is not discretionary — concrete, masonry, taping, and flooring all have temperature requirements, and the alternative to heating is stopping. The real lever on that line is not the budget; it is the enclosure date, worth $13,650 a week.
Sanitation and its placement, because the productivity arithmetic dwarfs the line item: floor-mounted units cost a fraction of the $1,848 a day you lose when 210 people walk four flights and 300 feet each way.
The general lesson is that temporary-facility cuts move money from an invisible account — craft productivity, which lands mostly on subcontractors — into a visible one, your general conditions, and then everyone congratulates themselves on the saving. Cut the lines whose failure mode you can actually see and price, and be honest that most of this budget is buying hours, not comfort.
18.7 Weather Management
Weather is the most-claimed and least-substantiated delay in construction. I have read dozens of weather claims. Most of them consist of a stack of daily reports that say "RAIN" and a demand for thirty days. They fail, and they deserve to.
What your contract actually says
Weather clauses vary enormously, so read yours before you need it. Common structures:
- An allowance of anticipated adverse weather days built into the contract time, with only days beyond the allowance being excusable.
- A comparison to historical normals from a defined source (typically the National Weather Service / NOAA station nearest the site) over a stated number of years.
- A threshold test — adverse weather that prevents work for more than a defined portion of a shift on a controlling activity.
Northgate's clause allows 18 non-working weather days within the 565-calendar-day contract time. Beyond 18, a day counts if adverse weather prevents work on a controlling (critical-path) activity for more than half of the scheduled shift. Weather days are excusable but non-compensable — you get time, not money. That last point is the one that surprises people: an excusable non-compensable day protects you from liquidated damages, but you still burn $5,150 a day of extended general conditions out of your own pocket.
⚖️ What the contract says. "Excusable" and "compensable" are two different words and they are not synonyms. Excusable means you are not in breach and not liable for liquidated damages. Compensable means somebody pays your extended costs. Most weather clauses grant the first and deny the second, which is a deliberate risk allocation: the owner absorbs the schedule, you absorb the money. Know which one you have before the first hurricane season, because it changes how much weather contingency belongs in your general conditions.
The forecast-driven decision
The single most consequential weather call on a job is: do we pour or not?
Here is how Margo and Jamal Foster actually make it, the afternoon before:
| Input | Threshold that changes the answer |
|---|---|
| Precipitation probability and timing | Rain during placement or before initial set is the problem; rain four hours after finishing usually is not |
| Air temperature — next 72 hours, not just tomorrow | Cold-weather concreting practice requires protecting fresh concrete above a minimum temperature for a defined period; ACI publishes recommended practices for both cold- and hot-weather concreting |
| Subgrade condition | You cannot place on frozen ground, and you cannot place on mud |
| Wind and evaporation rate | High wind plus low humidity plus warm concrete equals plastic shrinkage cracking, regardless of air temperature |
| Protection available on site | Do you actually have blankets, enclosures, and heaters staged tonight, or are they a phone call? |
| Cost of stopping vs. cost of a defect | See below |
The arithmetic of the call, for the level-four topping pour:
| Option | Cost |
|---|---|
| Pour and it goes wrong — plastic shrinkage cracking across 16,500 SF, evaluation by Caldwell Structural, epoxy injection and topping repair, plus delay to the floor covering | $40,000–$90,000 and unrecoverable schedule |
| Don't pour — 12-person crew redeployed (not sent home), pump truck cancelled inside 24 hr, 20 ready-mix loads cancelled, one day of float consumed | Pump cancellation $850 · ready-mix short-notice cancellation $1,100 · crew redeployment inefficiency ~4 MH × 12 × $58 = $2,784 · ≈ $4,734 |
Four thousand seven hundred dollars is cheap insurance against a forty-thousand-dollar defect. Superintendents who have been through a bad pour make this call quickly. Superintendents who have not, pour.
Protection, wind, and the record
Cold-weather protection costs money and takes planning: insulated blankets on fresh concrete, heated enclosures, heated mix water and aggregate at the plant, temporary heat inside the building, snow removal from decks before pours, and a fire watch anywhere you are running open-flame or indirect-fired heaters. Hot-weather protection costs money too: night pours, retarding admixtures, ice or chilled water in the mix, fogging and evaporation retarders, sunshades, and shifted start times.
Wind limits for crane operation are not a rule of thumb; they are on the load chart and in the manufacturer's operating manual, and OSHA's cranes standard for construction (29 CFR 1926, Subpart CC) requires operating within the manufacturer's procedures. Capacity is reduced as wind speed rises, and sail area matters more than weight. A 14,800-pound precast panel and a 6,200-pound curtain-wall bundle are both large flat objects that behave like sails; they come down long before the crane's own machine limit is reached. Two field realities: the wind at the boom tip is meaningfully stronger than the wind at the gate, so measure where the load is, and the operator has absolute authority to refuse a pick. If your site culture makes it hard for an operator to say no, you do not have a safety program; you have a slogan.
Protection of installed work is its own discipline: temporary roofing before the permanent membrane, dry-in of openings, protecting finished floors, covering ductwork ends so a wet-season shutdown does not put water in a system that will be commissioned in July, and keeping the building envelope closed once it is closed.
The record-keeping requirement. If you ever want to be paid or excused for weather, your daily report (Chapter 26) must contain, every single day, whether it rained or not:
- Temperature high and low, precipitation with amount, wind, and ground/site condition.
- Which specific activities were affected, by area — not "the site."
- Which crews were sent home or redeployed, by trade, with headcount and hours not worked.
- Why no other work was available for those crews. This is the one everybody omits and the one that decides the claim.
- Photographs with a visible date.
- The controlling activity that day, per the current schedule update.
Point 4 is the whole ballgame. An analyst reviewing your claim will ask: it rained, and you could not place concrete — but why did the drywall crew inside the dried-in south wing not work? If you have no answer, you have no weather day. This is exactly the entitlement/causation/damages structure you will meet in Chapter 33.
🔄 Check your understanding. Your contract gives excusable, non-compensable weather days. Twelve rain days beyond your allowance hit critical-path work. What do you get, what do you not get, and what is the dollar exposure?
Answer
You get 12 calendar days of time extension — which means no liquidated damages for those 12 days, saving 12 × $5,500 = $66,000 of LD exposure. You get no money, so you absorb 12 × $5,150 = $61,800 of extended general conditions yourself. Net: the clause protects you from about half of your daily exposure. And you get none of it unless your daily reports establish, day by day, which controlling activity was stopped and why no other work was available.
18.8 The Daily Rhythm
A well-run job has a heartbeat, and it is the same every day. Predictability is not rigidity; it is what lets two hundred people coordinate without meetings.
| Time | What happens | Why it exists |
|---|---|---|
| 5:45–6:15 | Superintendent's pre-gate walk, alone, in the dark | Find today's obstacles before anyone hits them. Fifteen minutes; see the hook. |
| 6:20 | Foremen's huddle at the trailer — 10 minutes standing, no chairs | Confirm today's plan, today's crane and hoist picks, today's inspections, today's conflicts. Standing keeps it to ten minutes. |
| 6:30 | Gate opens. Trade pre-shift huddles and stretch-and-flex, by crew, at the work face | Each foreman runs their own crew's task-specific hazard review. This is the JHA (job hazard analysis) made real — see Chapter 24. |
| 6:45–8:30 | The morning walk. Superintendent covers the building, floor by floor | Verify the plan is happening. Catch drift while it is cheap. |
| ~10:00 | The coordination touch — 12 minutes, at a table on the deck, with the three or four foremen whose work interferes today | Not a meeting. A touch. Resolve today's interferences before the afternoon. |
| 11:00–1:00 | Deliveries, inspections, and the paperwork window | AHJ inspections, delivery receiving, and the writing that has to happen. |
| 1:00–2:30 | The afternoon walk, quality and safety focused | Different eyes than the morning. Look at what was installed, not what is planned. |
| 2:30–3:00 | Tomorrow's look — with each foreman: what is ready, what is not, and why | The most valuable thirty minutes of the day. This is where standing still gets prevented. |
| 3:00–3:30 | Publish tomorrow's daily plan, crane pick schedule, and delivery schedule | In writing, posted, texted. Not "understood." |
| 3:30–4:15 | Close the site: crane secured, hoist locked, power down, gate locked, debris containers staged | Security, safety, and a clean start tomorrow. |
Superintendent versus project manager
In Chapter 1 you saw Ray's Tuesday and Margo's Tuesday laid side by side. Here is what the comparison actually teaches, distilled:
| Project manager (Ray) | Superintendent (Margo) | |
|---|---|---|
| Time horizon | Weeks to months | Today and tomorrow, with a six-week look-ahead |
| Primary medium | Written — email, RFIs, contracts, reports | Spoken and physical — conversations, walks, the plan on the wall |
| Interruption pattern | Interrupted from deep work; needs blocks | Interruption is the work; needs mobility |
| Unit of decision | Dollars and contract risk | Hours and access |
| Failure mode | Loses the field; makes decisions the field cannot execute | Loses the record; solves problems that later cannot be proven happened |
| What "good" looks like | The problem was seen four weeks out | The crew never stopped |
| Where they need each other | Ray needs Margo's early warning | Margo needs Ray's paper |
That last row is the whole partnership. Margo sees a problem at 6:19 a.m. that will cost $1,155 today. Ray sees a problem in a submittal log that will cost $244,950 in November. Neither can see the other's problem, and a job where those two people are not talking twice a day is a job that is going to hurt somebody's bottom line.
🪞 Learning check-in.
You are halfway through the book and at the front door of field operations, so let us be honest for a minute.
Go back through this chapter in your head — not the pages, just what you remember — and ask yourself three questions.
First: which sections did you skim? Most readers skim housekeeping and temporary facilities, because they read as janitorial. If that is you, notice that §18.4 and §18.6 contain two of the largest dollar figures in the chapter — a $388,000 cleanup budget and a $184,800 walk to the toilet — and that the double-handling arithmetic in §18.3.3 turned "put the drywall over there" into $18,616 plus three and a half days of the building's vertical capacity. What you skimmed is a map of what you currently believe is beneath your attention. That belief is the most expensive thing you own right now.
Second: when you pictured yourself in this chapter, where were you standing? In the trailer, or on the deck at 6:15? There is no wrong answer — Ray and Margo are both good at their jobs and neither could do the other's. But you should know which one you are, because the skill you will have to build deliberately is the other one. PMs who never walk the building make decisions the field cannot execute. Superintendents who never write anything down win arguments in the moment and lose them eighteen months later in a conference room.
Third: what is your instinct when a crew is standing still? Be honest. Most people's first instinct is to find out who is at fault. Notice that Margo's first instinct in every one of the five hook items was to find out what would remove the obstacle, and that fault-finding appeared exactly once — in a notebook column she reviews on Fridays, on purpose, later. Those are different reflexes and they produce different job sites. If your instinct is the first one, you can change it, but only if you notice it first.
Write down one sentence: the thing in this chapter I was most tempted to treat as beneath me. Come back to it after Chapter 27, which is this chapter with a theory attached.
18.9 Site Management for Special Conditions
The techniques above assume a fenced site with room to work. Many jobs are not that.
Occupied buildings and phased turnover
Northgate's north property line sits twenty feet from a Meridian clinic that stays open six days a week and takes ambulances. That single fact changes everything: crane swing paths, work hours, truck routes, noise, dust, parking, and the fire lane on Alder Street.
Rules that survive contact with reality:
- The occupant's operations are the constraint, not an inconvenience. Their loading dock hours, their ambulance route, and their patient arrival times are fixed points. You schedule around them.
- Never swing a load over an occupied area. Not "carefully." Never. Plan the pick radius so the load path stays inside your fence, and if it cannot, the work happens when the area is closed and barricaded, with a written lift plan.
- Phased turnover means phased everything — phased life safety, phased fire alarm, phased utilities, phased inspections, phased punch lists, and a written matrix showing exactly which areas are the owner's and which are yours on which date. The day an area becomes the owner's, your insurance, your security, and your housekeeping obligations change.
- One named person owns the interface with the occupant, and it is not whoever answers the phone.
Healthcare work and ICRA
If you build in or next to an operating healthcare facility, you will meet ICRA — Infection Control Risk Assessment. This is real, it is serious, and construction dust genuinely kills immunocompromised patients through fungal infection. Treat it accordingly.
How it works at a construction manager's level:
- The facility's infection prevention team — not the contractor — performs the assessment. It crosses the type of construction activity (from minor inspection through major demolition) with the risk group of the adjacent patient population (from office areas through immunocompromised and surgical areas) to produce a required precaution class.
- The precaution class dictates the barrier. Lower classes may allow dust control at the work point and sealed openings. Higher classes require hard barriers with a sealed anteroom, negative air pressure inside the work zone maintained by HEPA-filtered air scrubbers exhausting away from patient areas, tacky mats at the entrance, HEPA vacuuming, sealed debris carts, and dedicated routes and elevators for construction traffic and waste.
- Negative pressure gets monitored and documented daily, usually with a manometer at the barrier and a log that somebody signs. That log is a compliance record and it will be audited.
- Interim Life Safety Measures (ILSM) run in parallel whenever construction impairs fire alarm, sprinkler, smoke compartmentation, or egress. Extra fire watches, extra extinguishers, alternate exit signage, and daily inspection.
- The permit-and-inspection layer is different too: a healthcare facility typically has its own construction permit process on top of the local authority having jurisdiction (AHJ), and accreditation surveyors can appear unannounced.
The practical translation for you: ICRA is a logistics problem disguised as a medical one. Dedicated routes, dedicated elevators, sealed carts, and scheduled waste removal are all things this chapter already taught you. What ICRA adds is that the consequence of a mistake is not a backcharge. Build the barrier before you break the wall, document the pressure every day, and never, ever let a trade prop open an anteroom door because it is faster.
Urban sites with no laydown
Some jobs have zero square feet of storage. Everything is JIT by force. What changes:
- Delivery windows become appointments measured in minutes, often outside business hours, sometimes with a police detail and a street-closure permit.
- Material goes truck to crane to final position, always. There is no plan B, so the pick schedule must be right the first time.
- You buy space you do not own: sidewalk sheds, street occupancy permits, and sometimes off-site consolidation warehousing where material is received, sorted into installation kits, and shipped to the site in sequence. That warehousing is expensive and it is still usually cheaper than double handling.
- Prefabrication stops being a preference and becomes the plan. If you cannot store components, you buy assemblies. That is Chapter 39.
High-security and remote sites
High-security sites (data centers, defense, correctional, some utility work) add cleared-worker rosters, escorted access, tool and electronics inventories in and out, photography restrictions that break your normal documentation practice, and a badging process with lead time measured in weeks. Budget the productivity effect: a workforce that spends 25 minutes clearing security each morning and 15 each evening is losing roughly 40 minutes of an 8-hour shift — about 8% of your labor before anyone lifts a tool. Price that in the bid or eat it.
Remote sites invert the problem. Nothing is nearby: no supplier, no equipment dealer, no labor pool, no cell coverage, no hotel. Logistics becomes the dominant cost. You carry spare parts you would never carry in a city, you run a bigger material buffer on purpose, you may house and feed the workforce, and your delivery lead times are measured in weeks. The just-in-time logic of §18.3.2 flips entirely — on a remote site, stockpiling is correct, because the cost of a stopped crew you cannot re-supply for nine days dwarfs the cost of handling.
🔄 Check your understanding. Section 18.3.2 argued that on a congested site, just-in-time delivery usually beats stockpiling. Section 18.9 says that on a remote site, stockpiling is correct. Both are right. What is the underlying rule?
Answer
Buffer against the failure you cannot recover from. On a congested urban site the unrecoverable failure is handling and congestion — you physically cannot store it, and every square foot of storage buys you a second handling — while a missed truck is a phone call and a four-hour recovery, because the supplier is fifteen miles away. On a remote site the unrecoverable failure is resupply — a missing part is nine days out and the crew you are housing and feeding cannot work — while storage is nearly free because you have land.
It is the same rule as the curtain-wall comparison in §18.3.2: do not pick an extreme. Size the buffer to the specific failure that would hurt most, and be able to say out loud what that failure is. If you cannot name it, you do not have a logistics strategy — you have a habit.
18.10 The Interface with the Community
The neighbor is a stakeholder whether you invited them or not, and the neighbor who is ignored becomes a claimant.
Work hours and noise. Most jurisdictions restrict construction hours and set noise limits, and the enforcement mechanism is usually complaint-driven. Know your permitted hours cold, post them at the gate, and get any variance for an early-morning pour or a night crane pick in writing, in advance. A 5:00 a.m. pour start without a variance is a stop-work order at 5:20.
Dust and dirt. Track-out onto the public street is the most common complaint and the easiest to fix: a rock construction entrance, a street sweeper on call, wheel washing when the ground is wet, and water trucks or calcium chloride on haul roads. Budget it — Northgate carries dust control inside the $104,000 site-logistics line — and understand that most jurisdictions treat visible track-out and fugitive dust as a code violation, not a courtesy issue.
Truck routes. Publish them, enforce them, and tell your subcontractors and suppliers in writing. Every driver's navigation app will route them past an elementary school if you let it. A designated route with a map in every delivery confirmation is a five-minute administrative task that prevents the phone call from a city council member.
Complaints, and the neighbor who becomes a claimant. Run a complaint log the way you run a submittal log: date, time, name, contact, complaint, action taken, date closed, and who closed it. Respond to every complaint within one business day, even if the answer is "we cannot change that, here is why, and here is how long it lasts."
🏗️ From the field. On a mid-rise I ran years ago, a woman across the street called four times in three weeks about early-morning backup alarms. We logged it, said we would look into it, and did nothing, because the alarms are required equipment and we were within our permitted hours and we were right. Fourteen months later she was the lead complainant in a group that filed a nuisance and property-damage claim covering vibration, dust, and cracked plaster. The vibration claim was weak. The pattern of ignored complaints was not, and it is what made the file expensive. It cost us $31,000 and about sixty hours of my time. Four phone calls returned would have cost me forty minutes. Being right is not a strategy.
18.11 📋 Try It: Delivery Day at Northgate
The setup. It is a Thursday in late November, Year 1 — the day from the hook. You are the superintendent. You have one construction gate on Kirkbride Avenue, and its throat holds two trucks; a third stages in the street, which your city right-of-way permit prohibits between 7:00 and 9:00 a.m. because Kirkbride is a school route. Alder Street on the south is a fire lane only and must stay clear. You have one 90-ton crane and one hoist.
Fixed events you cannot move:
- Concrete pour, level-four deck topping, east half, 180 CY of lightweight structural topping at $206/CY delivered. Pump truck on the east drive from 5:45 to 11:30; the pump occupies half the gate throat. Ready-mix trucks arrive every 12 minutes from 6:30 to about 10:30 (20 loads). Jamal's placing and finishing crew is 12 people.
- Crane is at Pad A (north) until 9:10, then a 90-minute move to Pad C. The move crosses the east drive.
- AHJ inspection: Frank Petrosyan is coming at 9:00 for level-one in-wall rough-in. He does not wait, and he does not come back the same day.
- Hoist: 6:30–3:30, ~540 productive minutes.
Five delivery requests, all for the 7:00–11:00 window:
| # | Subcontractor | Load | Unloading method | Destination |
|---|---|---|---|---|
| 1 | Cordova Precast | 1 trailer, 3 panels (14,800 lb ea) | Crane at Pad A | Direct to wall — no laydown |
| 2 | Crossline Glass | 2 trailers, 40 curtain-wall units | Crane at Pad C | Floor staging L3–L4 south |
| 3 | Cardinal Mechanical | 1 box truck, duct and hangers | Hoist | L2 northeast — crew already there |
| 4 | Bellweather Interiors | 1 flatbed, studs + 60 sheets board | Telehandler | L1 mockup room |
| 5 | Halcyon Electric | 1 truck, conduit, strut, 2 panelboards | Hoist | L3 electrical room |
Your task. (a) Build the hour-by-hour delivery and crane schedule. (b) State what you reject or move, and to when. (c) Quantify what it costs if you simply let all five show up at 7:00.
Give it fifteen minutes with a pencil before you open the answer. Use these rates: laborer $52/MH · drywall/carpentry $68/MH · precast erector $81/MH · mechanical and electrical journeyman $89/MH · site blended $66/MH · crane all-in $310/hr · boom pump $295/hr.
Worked answer
Step 1 — Find the real constraints. There are four, and only four:
- The pump truck owns the east drive and half the gate until 11:30. Effective gate capacity in the morning is one truck at a time, not two.
- Ready-mix arrives every 12 minutes. Any truck in the throat when a mixer arrives puts the mixer in the street, which is both a permit violation and a discharge-time problem.
- The crane is at Pad A until 9:10 and then unavailable for 90 minutes.
- The hoist serves two competing deliveries (Cardinal and Halcyon) and its morning is already committed.
Step 2 — Classify each request against those constraints.
| # | Trade | Can it happen in the window? | Decision |
|---|---|---|---|
| 1 | Cordova Precast | Yes — crane is at Pad A, direct-to-wall, no laydown, no second handling | KEEP at 7:00. This is the one truck you cannot move without moving the crane plan. |
| 2 | Crossline Glass | No — crane is not at Pad C until 10:40, and two trailers cannot sit in the throat | MOVE. Trailer 1 to 10:40, trailer 2 to 1:00 p.m. |
| 3 | Cardinal Mechanical | Yes — hoist, feeds a crew already working on L2 | KEEP at 7:30, after the precast trailer clears. |
| 4 | Bellweather Interiors | No — needs the telehandler and drive space the pump owns | MOVE to 11:45, after the pump leaves. |
| 5 | Halcyon Electric | Partly — hoist is committed to Cardinal all morning | MOVE to 12:30. Panelboards need the hoist and a clear landing. |
Step 3 — The schedule.
| Time | Gate / deliveries | Crane | Hoist | Notes |
|---|---|---|---|---|
| 5:45 | Pump sets up, east drive | — | — | |
| 6:30 | Ready-mix begins, q12 min | Pad A — precast P-31→34 | Opens; Kestrel temp loads | Flagger at clinic drive |
| 7:00 | Cordova trailer in, unload direct | Erecting from the trailer | One truck only; mixers have priority in the queue | |
| 7:30 | Cardinal box truck in | Pad A — precast P-35→37 | Cardinal duct → L2 | Cordova trailer must be gone |
| 8:30 | Gate held for mixers | Cardinal continues | ||
| 9:00 | Frank Petrosyan — escort to L1 | Meet him at the gate. Walk him in. | ||
| 9:10 | CRANE MOVE A → C (90 min) | Crosses east drive between mixers — coordinate with the pump operator | ||
| 10:30 | Pour complete | Move continues | Cardinal wraps | |
| 10:40 | Crossline trailer 1 in | Pad C — curtain wall to L3–L4 | Direct to floor staging | |
| 11:30 | Pump demobs, drive clears | |||
| 11:45 | Bellweather flatbed in | Telehandler; staged in the mockup room, not the hoist lobby | ||
| 12:30 | Halcyon truck in | Halcyon → L3 | ||
| 1:00 | Crossline trailer 2 in | Pad C — curtain wall | ||
| 2:15 | Pad C — debris and pallets down | Scheduled, not optional | ||
| 2:45 | Contingency — held open |
Step 4 — What it costs if you let all five arrive at 7:00.
| Failure | Arithmetic | Cost |
|---|---|---|
| 2 ready-mix loads exceed discharge time in the queue and are rejected | 2 × 9 CY × $206/CY | $3,708 |
| Placing crew idle 45 min waiting for concrete | 12 × 0.75 hr × $58 | $522 |
| Boom pump standby | 0.75 hr × $295 | $221 |
| Crane sits 40 min at Pad A — precast trailer cannot get through the gate | 0.67 hr × $310 | $208 |
| Cordova erection crew (5) idle 40 min | 5 × 0.67 × $81 | $271 |
| Hoist queue: 22 mechanical and electrical workers on L2/L3 average 25 min waiting on material | 22 × 0.42 hr × $89 | $822 |
| Bellweather flatbed unloaded at the hoist lobby because the mockup room is unreachable — a guaranteed second handling of 60 sheets + studs | 6 MH × $52 | $312 |
| Right-of-way violation: trucks staged on Kirkbride in the school window | Citation plus a morning of the city's attention | not priced |
| Frank Petrosyan cannot get in, leaves, returns tomorrow — L1 in-wall cover-up slips one day | See below | see note |
| Priced subtotal | $6,064 |
The three unpriced items are the ones that matter.
- The cold joint. If the pour stalls past initial set you get an unplanned cold joint in a structural topping. Now Caldwell Structural has to evaluate it and either accept it or design a repair. That is an RFI, several days, and possibly a saw-cut-and-dowel detail. It is not on the list because you cannot price it in advance — which is exactly why you do not risk it for a delivery you could have moved with one phone call.
- The inspection. One day of slip on L1 cover-up costs $10,650 only if that work is on the critical path. If it has float, you did not spend $10,650 — you spent a day of float, which is a shared, consumable project asset that you will want back in March. Know which it is before you shrug. That distinction is Chapter 14.
- The relationship. Cordova, Cardinal, and Halcyon all lost time this morning because of a gate you did not manage. None of them will invoice you for it. All three of them now know that Northgate's delivery process does not work, and the next time you ask one of them for a favor on sequence, the answer will be different.
The lesson. Six thousand dollars of visible waste, a five-figure defect risk, a day of float, and three damaged relationships — all of it produced by five reasonable people each asking for a reasonable window, and one person who did not put the five requests on the same piece of paper. The paper is the job.
Spaced Review
Before you read each answer, try to produce it from memory. Retrieval is what makes this available at 6:15 in the morning, which is the only time it matters.
1. From Chapter 17 — the logistics plan as a living document. You drew a site logistics plan in preconstruction. How many versions should it have, and what triggers a new one?
Recall first. — One version per phase, minimum. The site during mass excavation, during foundations and steel, during enclosure, and during interior finishes is four different sites with four different gates, laydowns, crane positions, routes, and temporary facilities. A single logistics plan pinned to the trailer wall in March is a fiction by September. A new version is triggered by any change to the gate, the crane position, the laydown boundary, the hoist, the fire lane, or the phasing of an occupied area. This chapter is what happens when that document stops being a drawing and starts being a daily operating procedure: the delivery request form, the pick schedule, and the cleanup matrix are the plan in operation.
2. From Chapter 14 — where durations come from. A schedule activity says "hang drywall, level 3, 12 days." Where did the 12 come from, and what does §18.3 do to it?
Recall first. — Duration = quantity ÷ (crew size × production rate). The 12 days came from a quantity takeoff and a production rate: 94,000 SF at an assumed 11.5 man-hours per 1,000 SF is 1,081 man-hours, and 1,081 MH at a 12-person crew working 8-hour days is 11.3 work days — call it 12. Now connect it to this chapter: the production rate you assumed is a condition, not a property of the crew. Case Study 01 shows the same crew hanging the same board at 11.8 man-hours per 1,000 SF on a staged floor and 14.6 on an unstaged one. That is not a difference in skill or effort. It is a difference in whether material was where the work was. Your scheduled duration is a bet that somebody will create the conditions the production rate assumed. Site management is how that bet gets won.
3. Deep callback to Chapter 8 — the pour that stopped. At footing F-112, Ivy Kowalczyk said "don't pour this one" at 7:52 in the morning. Beyond the geotechnical problem, what did the next ninety minutes cost, and what does that teach you about idle time?
Recall first. — A nine-person crew, a pump truck, and six staged ready-mix trucks stopped. One loaded truck could be turned around for a short-load charge; one could not. The geotechnical repair was the headline — about $59,200 and three calendar days — but the quiet lesson is the one this chapter builds on: an idle crew is not a pause; it is a purchase. Nine people and a pump for ninety minutes is roughly $1,300 of nothing, and it happened for an excellent reason. Most idle time on your job will happen for a terrible reason — a blocked route, a missing pallet, a scaffold nobody moved, a truck at the wrong gate. Ivy's stop was worth every dollar. Margo's 6:19 a.m. scaffold, at $1,155, was worth none of it. Learning to tell those two apart, instantly, is most of what a superintendent knows.
Project Checkpoint: The Willow Street Site Logistics Plan — In Operation
In Chapter 17 you produced a site logistics plan for the Willow Street Community Center: a drawing (or a described plan) of your 2.1-acre site showing the gate, fence, trailer, laydown, crane position, temporary facilities, and truck route, with a version for each major phase. That was the plan. This chapter adds the operating layer — the five documents that turn a drawing into a way of running a day.
Recall the project: $6.8M, 24,000 SF, two stories, wood-framed second floor over a structural steel and CMU first floor, 425 calendar days, LDs $1,200/CD, municipal owner, prevailing wage, one existing 8-inch water main to relocate. Full package in Appendix K.
Produce five artifacts. Keep them to seven pages total.
1. A delivery scheduling procedure and form (1 page + a half-page procedure). Adapt the fields in §18.3.1. State the notice requirement (48 hours is standard; on a small site with one gate you may want 72), who receives the requests, how far in advance the schedule is published, and what happens to an unscheduled truck. Write that last sentence explicitly, because you will need to point at it in month four.
2. A lift plan and pick schedule format (1 page). Willow Street is two stories, so you will likely use a hydraulic truck crane on scheduled days rather than a standing crane, plus a telehandler. That makes crane days more precious, not less: on a crane day, every pick for every trade has to be on one sheet. Include the crane-move line item, a wind-check row, a debris-down pick, and a held-open contingency block.
3. Laydown assignments by trade and phase (1 page). Take your Chapter 17 site plan and overlay lettered laydown zones. Assign each zone to a trade for each phase, with square footage and the dates the assignment starts and ends. On 2.1 flat acres you have more room than Northgate does — which means your risk is not congestion, it is sprawl and double handling. Assign the space anyway.
4. A housekeeping and cleanup responsibility matrix (2 pages). One row per trade from your Chapter 16 buyout log. Columns: daily debris removal · sweeping · dust control · container and stream (metals, cardboard and clean wood, gypsum, landfill) · protection of installed work · pro-rata composite share · backcharge rate and the notice procedure. Every cell must be traceable to language in your subcontract scope sheet. Where a cell has no contract language behind it, you have just found a scope gap — mark it in red and write the clause.
5. A weather-protection plan with written triggers (1–2 pages). State your contract's weather-day definition and allowance. Then write the trigger conditions as if-then statements a superintendent can act on at 4:00 p.m. the day before: if the forecast low is below X within Y hours of a placement, then blankets are on site by end of shift and the pour decision is made by 3:00 p.m. Cover concrete placement, masonry, roofing, the wood-framed second floor (which is a water-damage exposure your steel-and-CMU first floor is not), crane wind limits, and protection of installed finishes. Include the daily-report weather fields from §18.7 — all six.
Why this matters for your notebook. These five documents are what an employer means when they ask whether you have "run a job." Anyone can draw a logistics plan. The delivery form and the cleanup matrix prove you understand that a plan without an operating procedure is a picture.
Next: Chapter 19 takes the other half of the field. You have now organized the site. Chapter 19 organizes the companies — prequalification criteria, the scope gaps between subcontracts, and the coordination structure that gets twenty trades to work in the same building without wrecking each other's work. Your cleanup matrix is the first draft of that coordination structure, so keep it handy.
Chapter Summary
The one sentence: A superintendent's job is to remove the reasons people are standing still.
The economics you should be able to reproduce from memory:
| Fact | Northgate number | Why you carry it |
|---|---|---|
| Site-wide labor burn, one hour at peak | 210 × $66 = $13,860 | One hour of site-wide idleness costs more than a day of being late |
| Daily schedule exposure | $5,150 + $5,500 = $10,650/CD | The comparison that makes the first row land |
| Double-handling all the gypsum board once | 8,585 sheets → 358 MH → $18,616 | Plus 3.6 days of hoist capacity you also spent |
| Temporary heat | $1,950/day | Every week of earlier enclosure is $13,650 |
| Toilets four floors away | 28 MH/day = $1,848/day | Amenity decisions are productivity decisions |
| Composite cleanup budget | $388,000 | Housekeeping is a line item, not a virtue |
The operating documents. If you have these five, you are running the job. If you do not, the job is running you.
- The delivery request form and published delivery schedule — 48-hour notice, destination required, unloading method required.
- The daily crane pick schedule — including the crane-move line, the debris-down pick, and a held-open contingency block.
- The hoist plan — capacity in minutes per day, allocated in advance, not by whoever reaches the call button first.
- The housekeeping and cleanup responsibility matrix — traceable, cell by cell, to subcontract language.
- The weather triggers — written as if-then statements, decided the afternoon before.
The decision framework for any logistics question:
Does this material have a DESTINATION (floor + zone)?
│ no ──▶ reject the delivery request until it does
│ yes
▼
Does the unloading method conflict with anything else in that window?
│ yes ──▶ resolve by priority: rolling trucks > life safety >
│ direct-to-install > crew already waiting > critical path >
│ most-recently-bumped
│ no
▼
Will it be handled more than once?
│ yes ──▶ price the second handling. If it exceeds the cost of
│ waiting for the right window, wait.
│ no
▼
Is the route to it 6 feet wide and clear? Is the floor rated for the load?
│ no ──▶ fix that first. Both answers are cheaper today.
▼
Schedule it. Publish it. Post it at the gate.
The three mistakes that cost the most, and the fix:
| Mistake | What it costs | Fix |
|---|---|---|
| Accepting a delivery with no destination | One extra handling, minimum — and hoist time you did not budget | Make "destination: floor and zone" a required field. Reject without it. |
| Letting the crane and the pour be scheduled by two different people | A crane move across a live pour, or a pick that never happens | One document. One person publishes it. Posted at the gate. |
| Backcharging cleanup without written notice | A disputed charge you will trade away for nothing | Notice, 24 hours to cure, rate stated in the subcontract. Then charge. |
What's Next
You have organized the ground. Now you have to organize the people standing on it — and almost none of them work for you. Chapter 19 is about the leverage you actually have over twenty companies whose employees you cannot direct: the subcontract, the schedule, and the coordination structure. It carries this book's hardest threshold concept, and it is the one that separates project managers who get results from ones who get frustrated. After that, Chapter 20 goes inside the production rates you have been assuming all chapter — where they come from, why they vary, and what actually moves them.