Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Magazine Supply Optimization: a Case-study

Duong Nguyen, Ana Ulianovici, Sami Achour, Soline Aubry, Nicolas Chesneau

TL;DR

The paper addresses magazine supply optimization under fixed inventory across a network of over 20,000 POS, where demand is irregular and affected by issue characteristics, holidays, and extra-products. It introduces AthenIA, a four-stage data-centric pipeline that combines SME-informed demand sensing with a novel Group Conformalized Quantile Regression (GCQR) approach and a scenario-based optimization engine to maximize profit while respecting practical constraints. Key contributions include the GCQR-driven demand sensing, scalable plate-level optimization over large retail networks, and industrial deployment with MLOps, cloud storage, and a user-friendly interface. Empirical results demonstrate profit improvements and reduced out-of-stock events alongside lower total supply, underscoring both economic and environmental benefits of more efficient magazine distribution.

Abstract

Supply optimization is a complex and challenging task in the magazine retail industry because of the fixed inventory assumption, irregular sales patterns, and varying product and point-of-sale characteristics. We introduce AthenIA, an industrialized magazine supply optimization solution that plans the supply for over 20,000 points of sale in France. We modularize the supply planning process into a four-step pipeline: demand sensing, optimization, business rules, and operating. The core of the solution is a novel group conformalized quantile regression method that integrates domain expert insights, coupled with a supply optimization technique that balances the costs of out-of-stock against the costs of over-supply. AthenIA has proven to be a valuable tool for magazine publishers, particularly in the context of evolving economic and ecological challenges.

Magazine Supply Optimization: a Case-study

TL;DR

The paper addresses magazine supply optimization under fixed inventory across a network of over 20,000 POS, where demand is irregular and affected by issue characteristics, holidays, and extra-products. It introduces AthenIA, a four-stage data-centric pipeline that combines SME-informed demand sensing with a novel Group Conformalized Quantile Regression (GCQR) approach and a scenario-based optimization engine to maximize profit while respecting practical constraints. Key contributions include the GCQR-driven demand sensing, scalable plate-level optimization over large retail networks, and industrial deployment with MLOps, cloud storage, and a user-friendly interface. Empirical results demonstrate profit improvements and reduced out-of-stock events alongside lower total supply, underscoring both economic and environmental benefits of more efficient magazine distribution.

Abstract

Supply optimization is a complex and challenging task in the magazine retail industry because of the fixed inventory assumption, irregular sales patterns, and varying product and point-of-sale characteristics. We introduce AthenIA, an industrialized magazine supply optimization solution that plans the supply for over 20,000 points of sale in France. We modularize the supply planning process into a four-step pipeline: demand sensing, optimization, business rules, and operating. The core of the solution is a novel group conformalized quantile regression method that integrates domain expert insights, coupled with a supply optimization technique that balances the costs of out-of-stock against the costs of over-supply. AthenIA has proven to be a valuable tool for magazine publishers, particularly in the context of evolving economic and ecological challenges.
Paper Structure (13 sections, 4 equations, 8 figures, 4 tables)

This paper contains 13 sections, 4 equations, 8 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Problem formulation
  3. AthenIA supply optimization solution
  4. Pipeline
  5. Human-in-the-loop demand sensing using group conformalized quantile regression
  6. Supply optimization
  7. Business rules and Operating
  8. Evaluation
  9. Evaluation of the demand sensing model
  10. Evaluation of the supply optimization method
  11. Deployment
  12. Related work
  13. Conclusion

Figures (8)

  • Figure 1: Histogram of a random subsetof the sales. The sales follow a typical long-tail distribution.
  • Figure 2: The supply optimization process is decomposed into four stages.
  • Figure 3: The irregularity in the sales of magazines.
  • Figure 4: Atypical sales. The circled peak was because that issue was sold with a particularly successful extra-product.
  • Figure 5: Example of the revenue and profit of different scenarios. Each dot corresponds to a scenario. See text for details.
  • ...and 3 more figures