Chapter 32: Further Reading — Inventory and Supply Chain Analytics

The resources here are annotated with honest assessments of what each offers and who will find it most useful. All books and papers cited are real, available publications.

Books

"Inventory Management Explained: A Focus on Forecasting, Lot Sizing, Safety Stock, and Ordering Systems" by David J. Piasecki The most practical book on inventory math for non-academic readers. Piasecki covers EOQ variants (for quantity discounts, backorders, and multiple products), safety stock calculations under different demand patterns, and continuous vs. periodic review systems. More technical than this chapter but accessible to readers who have worked through the chapter material. Highly recommended if you want to go deeper on the calculations without a full operations management textbook.

"The Goal: A Process of Ongoing Improvement" by Eliyahu M. Goldratt A business novel — an unusual format that works remarkably well. Goldratt's Theory of Constraints argues that every system has one primary constraint, and that optimizing anything other than that constraint produces little real improvement. Applied to supply chains, this reframes inventory management as a question about flow and throughput rather than cost minimization alone. Essential reading for understanding why the formulas in this chapter exist and what they are actually optimizing for.

"Supply Chain Management: Strategy, Planning, and Operation" by Sunil Chopra and Peter Meindl The widely adopted MBA-level textbook. Chopra and Meindl are especially strong on the strategic dimensions of supply chain design — network configuration, the tradeoff between responsiveness and efficiency, risk pooling across locations — that this chapter only touches. The sections on demand uncertainty and the value of information are directly relevant to understanding why better data (your Python dashboard) translates directly to better supply chain performance.

"Lean Thinking" by James P. Womack and Daniel T. Jones Lean manufacturing principles, extended from Toyota's production system to supply chains and services. The core insight — that waste (excess inventory, unnecessary steps, waiting time) is embedded in every process and can be systematically eliminated — provides a valuable lens for interpreting supply chain analytics. When your analysis reveals $85,000 in slow-moving inventory, "Lean Thinking" explains why that happened and how to prevent recurrence.

"Operations Management" by William Stevenson The standard undergraduate operations management textbook. Chapters on inventory management, aggregate planning, and demand forecasting provide the formal academic treatment of the formulas introduced in this chapter, with derivations and worked examples. If you want to understand where the EOQ formula comes from mathematically, or why the combined variance formula for safety stock takes the form it does, Stevenson is the right reference.

Articles and Papers

"The Bullwhip Effect in Supply Chains" by Hau L. Lee, V. Padmanabhan, and Seungjin Whang, Sloan Management Review, 1997 A landmark paper explaining why small fluctuations in customer demand amplify into large swings in orders as you move upstream in the supply chain. Understanding the bullwhip effect is essential for correctly interpreting supply chain data — it explains why your purchase orders to suppliers should not mirror your customer demand patterns, and why supplier lead time variability is often worse than raw data suggests. Available through most business school library databases.

"A Review of Inventory Management Research in Major Logistics Journals" — Multiple authors There are several solid review articles in journals like the International Journal of Production Economics and the European Journal of Operational Research that survey the academic literature on inventory optimization. These are useful if you want to understand the boundaries of what the formulas in this chapter can and cannot do, and where more sophisticated approaches (multi-echelon optimization, stochastic programming) become necessary.

APICS/ASCM Operations Management Body of Knowledge The Association for Supply Chain Management publishes the APICS Dictionary and supporting reference material that defines standard industry terminology precisely. When you need to communicate with supply chain professionals using the correct vocabulary, the APICS reference material is authoritative. Available at ascm.org.

Online Resources

MIT OpenCourseWare: 15.762 Supply Chain Planning Free access to MIT's graduate supply chain management course materials, including lecture notes and problem sets on inventory optimization, demand forecasting, and distribution network design. Available at ocw.mit.edu. The problem sets are substantially more mathematically demanding than this chapter but are valuable for readers who want graduate-level treatment of the material.

ASCM (formerly APICS) Learning Center The Association for Supply Chain Management offers free articles, webinars, and e-learning modules on supply chain topics at ascm.org. The CPIM (Certified in Planning and Inventory Management) certification curriculum maps directly to the concepts in this chapter and represents the professional standard for operations and inventory management practitioners.

Gartner Supply Chain Top 25 Annual Report Gartner's annual ranking of the world's leading supply chains is accompanied by analysis of what capabilities distinguish excellent supply chain organizations from average ones. Analytics and data capabilities are consistently cited as primary differentiators. Available at gartner.com. Useful for understanding the strategic context of the technical work in this chapter.

Institute for Supply Management (ISM) ISM publishes the monthly PMI (Purchasing Managers Index) and a range of practitioner-focused articles on supplier management, procurement practices, and supply chain risk. Relevant for understanding supplier performance management in business context. Available at ismworld.org.

Python Libraries for Going Deeper

statsmodels You encountered ExponentialSmoothing from statsmodels in this chapter. The library also contains ARIMA, SARIMA, and state-space models for more sophisticated demand forecasting. For Class A items with pronounced seasonal patterns or trend behavior, these models can meaningfully improve forecast accuracy — which directly reduces required safety stock. Documentation at statsmodels.org.

prophet (developed by Meta, open source) A forecasting library designed specifically for business time series with multiple seasonality levels, holiday effects, and occasional changepoints. If your business has Black Friday spikes, seasonal product lines, or other complex demand patterns, Prophet often outperforms exponential smoothing with less parameter tuning. Available on PyPI as prophet. Documentation at facebook.github.io/prophet.

scipy.optimize The optimization routines in SciPy can be used to extend the EOQ model to more realistic scenarios: quantity discounts, minimum order quantities, joint replenishment across multiple SKUs from the same supplier. scipy.optimize.minimize_scalar and minimize handle both univariate and multivariate optimization. Documentation at scipy.org.

scikit-learn Chapter 34 covers scikit-learn in depth, but it is worth noting here that many supply chain problems can be framed as machine learning problems: predicting whether a supplier will be late based on historical patterns, forecasting demand with regression models, or classifying orders as high/low risk of stockout. The same library you will use for customer churn prediction has direct supply chain applications.

pyomo For production-scale supply chain optimization problems — multi-echelon inventory optimization, network design, transportation planning — PyPI's pyomo package provides a full optimization modeling language for Python. Requires understanding of linear and mixed-integer programming to use effectively, but is the tool of choice for industrial-strength supply chain optimization problems.

Connecting to Other Chapters in This Book

Chapter 23 (Working With Databases): The acme_inventory.db SQLite database used throughout this chapter was introduced in Chapter 23. The skills you built there — designing schemas, writing queries, using parameterized SQL — are the foundation of the inventory analytics work here. If you want to extend this system to handle live data updates or more complex queries, revisit Chapter 23's database design section.

Chapter 26 (Time Series Forecasting): Demand forecasting for inventory is a direct application of Chapter 26's exponential smoothing and seasonal decomposition techniques. The connection is precise: better forecasting accuracy means smaller demand variance, which means less safety stock required, which means lower working capital costs. Investing in Chapter 26 techniques pays dividends in supply chain working capital.

Chapter 33 (Interactive Dashboards with Streamlit): The Excel dashboard built in this chapter is functional but not interactive — stakeholders cannot filter by region or drill into specific suppliers without help. Chapter 33 shows how to convert the exact same underlying Python calculations into a live, browser-accessible dashboard with filter controls and automatic data refresh.

Chapter 34 (Predictive Models): Supply chain data is rich territory for predictive modeling. You can predict whether a supplier will deliver late based on historical patterns, forecast whether a SKU will stock out within 14 days, or identify which new products are likely to become Class A items. The machine learning workflow from Chapter 34 applies directly to supply chain prediction problems.