Comparison of Optimization Techniques for Complex Supply Chain Network Planning Problems
de Araújo Pessoa Luís Filipe, Horstkemper Dennis, de Siqueira Diego, Hellingrath Bernd, Gomes Pereira de Lacerda Marcelo, Buarque de Lima Neto Fernando
Planning tasks in Supply Chain Management demonstrate a high complexity due to several influences such as globalization effects, mass customization of products and shorter product life cycles. This is leading to additional requirements for logistics planning in general. Nevertheless, the planning and management of supply chains is a key capability for today’s companies in order to increase profitability as well as decrease the uncertainty and the risk of certain actions taken by one or more members of the supply chain. In this context, Supply Chain Network Planning (SCNP) is used to determine an adequate allocation of demands and capacities in storage, transportation and production facilities, aiming to fulfill customer demands at minimal costs. As a planning model on the tactical level, SCNP can be used as an important input for operational planning tasks, e.g. for a more detailed transport planning. However, due to the complexity of the problem, exact mathematical solution techniques can consume a high amount of time, and are as such not applicable in practice. Logistics decision support systems, like Advanced Planning Systems (APS) extend them therefore with heuristics or offer meta-heuristic solution approaches, each in order to provide good solutions in short timeframes. Despite this, there is a constant need for new, more efficient techniques to tackle the increasing complexity of modern supply chains and to handle ever shorter planning cycles. Therefore, this paper seeks to explore the possibilities of applying modern meta-heuristic optimization techniques onto SCNP problems.
Supply Chain Planning; Optimization; Metaheuristics