From data collection to evaluation and visualization of prediction results, this book provides a comprehensive overview of the process of predicting demand for retailers. Each step is illustrated with the relevant code and implementation details to demystify how historical data can be leveraged to predict future demand. The tools and methods presented can be applied to most retail settings, both online and brick-and-mortar, such as fashion, electronics, groceries, and furniture. This book is intended to help students in business analytics and data scientists better master how to leverage data for predicting demand in retail applications. It can also be used as a guide for supply chain practitioners who are interested in predicting demand. It enables readers to understand how to leverage data to predict future demand, how to clean and pre-process the data to make it suitable for predictive analytics, what the common caveats are in terms of implementation and how to assess prediction accuracy.

"Demand Forecasting is an important tool in many industries including retail. Althoughmany approaches have been developed to accurately predict the demand of productsbased on their historical sales data, demand prediction is still a complex issue especiallywhen there is a domain shift between training and testing data.In this work, we study three examples of domain shifts in the context of retail: outbreak ofthe COVID-19 pandemic, opening a new store, and introducing a new product. We firstshow that the accuracy of demand prediction models suffers after each sudden change.Then, we use domain adaptation methods, such as Frustratingly Easy (FE) and KernelMean Matching (KMM) to help improve the demand prediction accuracy by leveragingthe available data from the period before the shift (source domain) and adapting it to thedata after the shift (target domain). Additionally, we show that using a pairing techniquefurther helps improve the prediction accuracy.We use two methods as our base forecasting model: XGBoost and Transformers, and weshow that in the context of our data, it is better to use XGBoost.Our dataset comprises of point-of-sales data from 89 locations of Alimentation Couche-Tard convenient stores in the island of Montreal gathered between 2019-07 and 2021-02.We use product price information in addition to sales information to predict the demandof products in each store. In this study, we focus our attention on the two high-sellingcategories of coffee and energy drinks"--

INTERMITTENT DEMAND FORECASTING The first text to focus on the methods and approaches of intermittent, rather than fast, demand forecasting Intermittent Demand Forecasting is for anyone who is interested in improving forecasts of intermittent demand products, and enhancing the management of inventories. Whether you are a practitioner, at the sharp end of demand planning, a software designer, a student, an academic teaching operational research or operations management courses, or a researcher in this field, we hope that the book will inspire you to rethink demand forecasting. If you do so, then you can contribute towards significant economic and environmental benefits. No prior knowledge of intermittent demand forecasting or inventory management is assumed in this book. The key formulae are accompanied by worked examples to show how they can be implemented in practice. For those wishing to understand the theory in more depth, technical notes are provided at the end of each chapter, as well as an extensive and up-to-date collection of references for further study. Software developments are reviewed, to give an appreciation of the current state of the art in commercial and open source software. “Intermittent demand forecasting may seem like a specialized area but actually is at the center of sustainability efforts to consume less and to waste less. Boylan and Syntetos have done a superb job in showing how improvements in inventory management are pivotal in achieving this. Their book covers both the theory and practice of intermittent demand forecasting and my prediction is that it will fast become the bible of the field.” —Spyros Makridakis, Professor, University of Nicosia, and Director, Institute for the Future and the Makridakis Open Forecasting Center (MOFC). “We have been able to support our clients by adopting many of the ideas discussed in this excellent book, and implementing them in our software. I am sure that these ideas will be equally helpful for other supply chain software vendors and for companies wanting to update and upgrade their capabilities in forecasting and inventory management.” —Suresh Acharya, VP, Research and Development, Blue Yonder. “As product variants proliferate and the pace of business quickens, more and more items have intermittent demand. Boylan and Syntetos have long been leaders in extending forecasting and inventory methods to accommodate this new reality. Their book gathers and clarifies decades of research in this area, and explains how practitioners can exploit this knowledge to make their operations more efficient and effective.” —Thomas R. Willemain, Professor Emeritus, Rensselaer Polytechnic Institute.

How should retailers leverage aggregate (category) sales information for individual product demand prediction? Motivated by inventory risk pooling, we develop a new prediction framework that integrates category-product sales information to exploit the benefit of pooling. We propose to combine data from different aggregation levels in a transfer learning framework. Our approach treats the top-level sales information as a regularization for fitting the bottom-level prediction model. We characterize the error performance of our model in linear cases and demonstrate the benefit of pooling. Moreover, our approach exploits a natural connection to regularized gradient boosting trees that enable a scalable implementation for large-scale applications. Based on an internal study with JD.com on more than 6,000 weekly observations between 2020 and 2021, we evaluate the out-of-sample forecasting performance of our approach against state-of-the-art benchmarks. The result shows that our approach delivers superior forecasting performance consistently with more than 9% improvement over the benchmark method of JD.com. We further validate its generalizability on a standard public data set. Our result highlights the value of transfer learning to demand prediction in retail with both theoretical and empirical support. Based on a conservative estimate of JD.com, the improved forecasts can reduce the operating cost by 0.01-0.34 RMB per sold unit on its platform, which implies significant cost savings for the low-margin e-retail business.

Using data science in order to solve a problem requires a scientific mindset more than coding skills. Data Science for Supply Chain Forecasting, Second Edition contends that a true scientific method which includes experimentation, observation, and constant questioning must be applied to supply chains to achieve excellence in demand forecasting. This second edition adds more than 45 percent extra content with four new chapters including an introduction to neural networks and the forecast value added framework. Part I focuses on statistical "traditional" models, Part II, on machine learning, and the all-new Part III discusses demand forecasting process management. The various chapters focus on both forecast models and new concepts such as metrics, underfitting, overfitting, outliers, feature optimization, and external demand drivers. The book is replete with do-it-yourself sections with implementations provided in Python (and Excel for the statistical models) to show the readers how to apply these models themselves. This hands-on book, covering the entire range of forecasting—from the basics all the way to leading-edge models—will benefit supply chain practitioners, forecasters, and analysts looking to go the extra mile with demand forecasting.

Incorporating 25 years of sales forecasting management research with more than 400 companies, Sales Forecasting Management, Second Edition is the first text to truly integrate the theory and practice of sales forecasting management. This research includes the personal experiences of John T. Mentzer and Mark A. Moon in advising companies how to improve their sales forecasting management practices. Their program of research includes two major surveys of companies′ sales forecasting practices, a two-year, in-depth study of sales forecasting management practices of 20 major companies, and an ongoing study of how to apply the findings from the two-year study to conducting sales forecasting audits of additional companies. The book provides comprehensive coverage of the techniques and applications of sales forecasting analysis, combined with a managerial focus to give managers and users of the sales forecasting function a clear understanding of the forecasting needs of all business functions. New to This Edition: The author′s well-regarded Multicaster software system demo, previously available on cassette, has been updated and is now available for download from the authors′ Web site New insights on the critical area of qualitative forecasting are presented The results of additional surveys done since the publication of the first edition have been added The discussion of the four dimensions of forecasting management has been significantly enhanced Significant reorganization and updating has been done to strengthen and improve the material for the second edition. Sales Forecasting Management is an ideal text for graduate courses in sales forecasting management. Practitioners in marketing, sales, finance/accounting, production/purchasing, and logistics will also find this easy-to-understand volume essential.

We study how retailers can use data aggregation and clustering to improve demand prediction. High accuracy in demand prediction allows retailers to more effectively manage their inventory and mitigate stock-outs and excess supply. A typical retail setting involves predicting demand for hundreds of items simultaneously. Although some items have a large amount of historical data, others were recently introduced, and thus, transaction data could be scarce. A common approach is to cluster several items and estimate a joint model at the cluster level. In this vein, one can estimate some model parameters by aggregating the data from several items and other parameters at the individual-item level. We propose a practical method referred to as data aggregation with clustering (DAC), which balances the tradeoff between data aggregation and model flexibility. DAC allows us to predict demand while optimally identifying the features that should be estimated at the (i) item, (ii) cluster, and (iii) aggregate levels. We show that the DAC algorithm yields a consistent estimate, along with improved asymptotic properties relative to the decentralized method, which estimates a different model for each item. Using both simulated and real data, we illustrate DAC's improvement in prediction accuracy relative to common benchmarks. Interestingly, the DAC algorithm has theoretical and practical advantages and helps retailers uncover meaningful managerial insights.