Abstract:The real-time joint optimization of inventory replenishment and vehicle routing is essential for cost-efficiently operating one-warehouse, multiple-retailer systems. This is complex, as future demand predictions should capture correlation retailer demand, and based upon such predictions, replenishment and routing decisions must be taken. Traditionally, such decisions are made by either making distributional assumptions or using machine-learning-based point forecasts. The former approach ignores nonstationary demand patterns, while the latter approach only provides a point forecast ignoring the inherent forecast error. Consequently, in practice, service levels often do not meet their targets, and truck fill rates fall short, harming the efficiency and sustainability of daily operations. We propose Scenario Predict-then-Optimize. This fully data-driven approach for online inventory routing consists of two subsequent steps at each real-time decision epoch. The scenario-predict step exploits neural networks, specifically multi-horizon quantile recurrent neural networks, to predict future demand quantiles, upon which we design a scenario sampling approach. The subsequent scenario-optimize step then solves a scenario-based stochastic programming approximation. Results show that our approach outperforms the classic Predict-then-Optimize paradigm and empirical sampling methods. We show this both on synthetic data and real-life data. Our approach is appealing to practitioners. It is fast, does not rely on distributional assumptions, and does not face the burden of single-scenario forecasts. We show it is robust for various demand and cost parameters, enhancing the efficiency and sustainability of daily replenishment and routing decisions. Finally, Scenario Predict-then-Optimize is general and can be easily extended to account for other operational constraints, making it a useful tool in practice.

Integrated Inventory and Transportation Management with Stochastic Demands: A Scenario-Based Economic Model Predictive Control Approach

Integration of production planning and scheduling under demand and utility uncertainties

Model Predictive Control with Dynamic Pricing and Probability Inventory of A Single Supply Chain Unit

A Chance Constrained Programming Approach for Multi-Product Multi-Stage Integrated Production Planning under Internal and External Uncertainties

Deep Reinforcement Learning Approach for Capacitated Supply Chain optimization under Demand Uncertainty

Model Predictive Control with an On-Line Identification Model of a Supply Chain Unit

A Progressive Hedging-Based Solution Approach for Integrated Planning and Scheduling Problems under Demand Uncertainty

Integrated inventory management and supplier base reduction in a supply chain with multiple uncertainties.

Integrated optimisation for production capacity, raw material ordering and production planning under time and quantity uncertainties based on two case studies

A multi-stage stochastic inventory management model for transport companies including several different transport modes

A Multi-Scenario Optimization Model for Emergency Cold Chain Logistics Distribution

Scenario Predict-then-Optimize for Data-Driven Online Inventory Routing

Joint Service Capacity Planning and Dynamic Container Routing in Shipping Network with Uncertain Demands

The replenishment policy of agri-products with stochastic demand in integrated agricultural supply chains

A Reliable Model for Integrated Supply Chain Design under Disruption Risk

A Multi-objective Mixed-integer Programming Approach for Supply Chain Disruption Response with Lead-Time Awareness

Integrated Planning and Allocation: A Stochastic Dynamic Programming Approach in Container Transportation

Data-driven robust optimization of dual-channel closed-loop supply chain network design considering uncertain demand and carbon cap-and-trade policy

Simulation-based Optimisation Approach for the Stochastic Two-Echelon Logistics Problem.

Algorithmic Approaches to Inventory Management Optimization

Economic Model Predictive Control for Transport-Reaction Systems with Target Profiles