Integrated MILP-Based Optimization and Flexibility Assessment of a Digitally Supported Postponement Logistics System

Abstract

Manufacturing companies are increasingly required to cope with a wider product mix, changing production volumes, and shorter response times. These challenges are particularly relevant in postponement environments, where final product differentiation takes place close to the actual customer demand and therefore places considerable pressure on internal logistics. This paper presents a mixed-integer linear programming (MILP) model developed for the operational optimization of a digitally supported postponement logistics system. The model links production-order assignment with the execution of the related intralogistics tasks and considers technological feasibility, milk-run operations, task sequencing, resource availability, timing requirements, service levels, and line-side storage limitations. The optimization is carried out in two stages. First, the maximum production volume that can be served with the available resources is determined. In the second stage, the logistics performance is improved while keeping the achievable volume close to this optimum. Besides supporting operational decisions, the model is also used to evaluate logistics flexibility. For this purpose, five indicators are defined, covering on-time task completion, logistics response time, mix flexibility, volume flexibility, and the stability of logistics performance under different operating conditions. The proposed approach therefore provides a common framework for analysing operational performance and flexibility and can support the comparison of alternative resource configurations and production scenarios in postponement logistics systems.