Introduction to Machine Learning for Supply Chains and Industrial Optimization

The Checklist Model of AI Maturation, Python, and Real-World Datasets
Erscheinungsjahr 2027
ISBN: 978-3-032-44429-5
Verlag: Springer

Buch, Englisch, 355 Seiten, Format (B × H): 155 mm x 235 mm

Reihe: Forum for Interdisciplinary Mathematics

Many machine learning books survey techniques one at a time. This book shows how they fit together. Its checklist model of AI maturation provides a unifying path from regression and experimental design through decision trees, neural networks, large language models, and clustering to genetic algorithms, Markov decision processes, partially observable Markov decision processes, and reinforcement learning.

A connected family of supply chain cases runs throughout the book, covering demand forecasting, supplier classification, vehicle routing, production scheduling, and inventory management. Each case includes printed “known answers,” allowing readers to check their results. Rather than relying on long program listings that can quickly become outdated, the book provides regeneration prompts that readers can use with LLM assistants such as ChatGPT, Gemini, and Claude to generate current Python implementations. Verification checklists and guidance for running code in Google Colab or locally help readers evaluate the generated programs rather than treating their output as automatically correct.

Designed explicitly for the era of LLM-assisted coding, the book combines statistical foundations, modern machine learning, and decision-focused optimization. End-of-chapter problems include LLM workflow exercises in which readers generate, run, and verify their own implementations, while solutions to selected problems appear at the back of the book.

A glossary, alternative plans for semester-long and shorter courses, real-world datasets, and supplementary teaching materials make the book adaptable for classroom and professional use. It is written for senior undergraduates, graduate students, researchers, and practitioners in supply chain management, industrial engineering, operations research, analytics, and related fields who have a working knowledge of basic statistics.

Zielgruppe

Graduate

Autoren/Hrsg.

Allen, Theodore T.

Liu, Enhao

Fachgebiete

• Mathematik | Informatik Mathematik Stochastik
• Wirtschaftswissenschaften Wirtschaftswissenschaften Wirtschaftswissenschaften: Allgemeines
• Mathematik | Informatik EDV | Informatik Professionelle Anwendung
• Wirtschaftswissenschaften Betriebswirtschaft Management
• Wirtschaftswissenschaften Betriebswirtschaft Wirtschaftsmathematik und -statistik
• Wirtschaftswissenschaften Betriebswirtschaft Unternehmensforschung

Inhaltsverzeichnis

Introduction.- Supervised Data and Continuous Outputs: Regression, The Prototypical Modeling Method.- Supervised Data and Continuous Outputs: Machine Learning Alternatives to Regression.- Supervised Data and Categorical Outputs: Logistic Regression and Support Vector Machines.- Experimental Design: Even with Supervised Data, There is Rarely Enough.- Categorical Unsupervised Data and Clustering.- Single Period Optimization: Genetic Algorithms.- Multi-Period Optimization: Markov Decision.- Multi-Period Optimization with Unknown State: Partially Observable MDPs.- Multi-Periods with Unknown Models: Bayesian Reinforcement Learning.

Über Autor(innen)

Theodore T. Allen is Professor of Mechanical and Industrial Engineering at the University of Massachusetts Amherst. Before joining UMass Amherst, he spent 30 years at The Ohio State University as a lecturer and as Assistant Professor, Associate Professor, and Professor of Integrated Systems Engineering. An internationally recognized expert in analytics, decision science, and machine learning, he has authored more than 90 peer-reviewed publications, including articles in , , and . He has served as President of the INFORMS Social Media Analytics Section, is a member of the INFORMS Edelman Academy, was an inaugural member of the MIT Election Data and Science Lab and is a Fellow of the American Society for Quality. Known for his engaging teaching style, he has received six student-selected teaching awards.

Enhao Liu is a Senior Healthcare Data Scientist at the University of Maryland Medical System, where he develops and deploys predictive, prescriptive, and simulation models supporting patient flow, discharge planning, and capacity management across the hospital system. He also architects real-time electronic medical record data pipelines and production AI systems. He earned his Ph.D. from The Ohio State University, specializing in operations research, and his bachelor’s degree in engineering from Jinan University in China. His research spans machine learning, model-based reinforcement learning, and partially observable Markov decision processes for cybersecurity maintenance, with publications in and .