English Version

首页» 学术科研» 讲座信息

[2014/07/04]Seminar at Peking University

Seminar at Peking University – 8 July 2014


主持人:侍乐媛  教授

时间:201478 上午 10:00



I will present some of my research and introduce my university and department.

In-plant logistics modeling for complex assembly systems

In order to have an efficient assembly system, a reliable material delivery system is indispensable. Given the current trend of ever more customized products, not only the assembly tasks themselves become more complex, but the in-plant logistics also become increasingly complicated. A multitude of variant parts has to be supplied, leading to many internal logistics flows and a crowded line side. To be able to deal with the increasing number of variants, there is a shift from bulk feeding all parts to the line towards other part feeding systems such as sequencing and kitting.

The choice for a specific part feeding system has multiple impacts on the internal transport flows, on the walking and searching of the assembly operator and on the additional material handling needed for kit preparation. In this talk, we will examine a mathematical model quantifying the costs related to the different impact factors. We consider assembly on mixed-model assembly lines, with a high variation between the models. A wide diversity of parts – ranging from very small to large and voluminous, and from common parts to low usage variant parts – is taken into account. Parts are delivered to the line by forklift transport (pallets) and milk runs (boxes and kits). Walking distances of the assembly operator at a work station are assumed to be proportional to the space occupation of the parts, which in turn depends on the parts’ feeding system. Despite this interdependence, through a reformulation the model was retained linear.

Optimization results show that no single material delivery system outperforms another in all situations. Instead hybrid policies, where some parts will be kitted and sequenced while others will be stocked in bulk at the line, are preferred. An analysis is done to study the factors influencing the preferred delivery method for individual parts. To this aim, an algorithm is developed for the creation of representative datasets. Parts are diversified according to weight, volume, usage rate, and so on, to explore the most critical characteristics for the material supply decision. The hypotheses tested, and corresponding results, will be discussed.



Prof. Dr. Veronique Limère

Faculty of Engineering and Architecture – Department of Industrial Management

Assistant Professor with research interest in facility logistics, material handling and operations research applications in production and logistics systems in general.

Check my profile on Linkedin

Email: veronique.limere@ugent.be