ABSTRACT

This paper first summarises a survey of logistics and production planning engineers in the automotive industry on the use of simulation models to solve planning tasks in logistics and production conducted by the authors. As a solution for the analysed trade-off between a required quick provision of simulation results and a tolerable inaccuracy of the simulation results, the authors describe and evaluate the application of mesoscopic discrete-rate-based simulation models for two typical production and logistics planning tasks in the automotive industry. In terms of level of detail, mesoscopic simulation models fall between object-based discrete-event simulation models and flow-based continuous simulation models. Mesoscopic models mainly represent logistics flow processes on an aggregated level through piecewise-constant flow rates by applying the discrete-rate simulation paradigm instead of modelling individual flow objects. This leads to fast model building and computation.

摘要

本文首先总结了作者对汽车行业物流和生产规划工程师使用仿真模型解决物流和生产规划任务的调查。作为在所需的快速提供仿真结果和可容忍的仿真结果不准确性之间进行分析权衡的解决方案，作者描述和评估了基于细观离散速率的仿真模型在两个典型的生产和物流计划任务中的应用在汽车行业。在详细程度方面，介观仿真模型介于基于对象的离散事件仿真模型和基于流的连续仿真模型之间。细观模型主要通过应用离散速率模拟范式而不是对单个流动对象进行建模，通过分段恒定流速来表示聚合级别的物流流程。这导致快速的模型构建和计算。