Data-driven optimization for automated warehouse operations decarbonization

Abstract

The rapid development of intelligent warehouse systems is resulting in the realization of automation in warehouse activities and raising awareness of decarbonization, particularly the need to reduce carbon emissions from electricity consumption. Driven by the decarbonization trend, microgrid systems with rooftop photovoltaic panels are becoming more popular in warehouses and are providing zero-carbon electricity for warehouse operations. How to make better use of microgrid systems and reduce the consumption of electricity generated from traditional energy sources is becoming increasingly important in warehouse systems. This paper investigates an operational problem in a warehouse system equipped with a shuttle-based storage and retrieval system, in which a microgrid system acts as the main electricity source. Power-load management is applied to avoid peaks of energy consumption, and a mixed linear programming model is developed to optimize task sequencing and scheduling with decarbonization awareness. To solve the proposed problem, a data-driven variable neighbourhood search algorithm is built. Numerical experiments are conducted to validate the model and algorithm. Sensitivity analysis shows the effectiveness of power-load management and the influence of system configuration on energy consumption.