Supply chain business modelling

• Main Author: Pereira, Gabriela
• Other Authors: Shah, Nilay
• Published: Imperial College London 2012
• Subjects: 660
• Online Access: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.616725

The developed work is motivated by the hypothesis that the presented Supply Chain Business Model is a practical and comprehensive approach to support not only operational day-to-day business decisions, but most importantly strategic and long term decisions that may define the success and the longevity of a business. Conceptually, the Business Supply Chain Model developed in this thesis replicates the behaviour and decision making of the different agents in a supply chain, and an Optimisation Module determines the optimised parameters that maximise the overall business profit, whatever scenario it may be. In the optimisation module, a Genetic Algorithm was used to determine the best equation parameters for each individual agent that optimise the overall supply chain profit. Furthermore, several business case-scenarios are presented and the findings highlighted. These case-scenarios prove that: the HC model is robust when subjected to predictable or unpredictable causes of variability; the bullwhip effect can be reduced significantly by applying GA as the optimisation tool; the improvement of profits needs to be evaluated at a global scale, independently of the individual agents’ profit; impact of supply shortages in the SC ; retail expansion analysis; delivery patterns change impact in profitability; impact of sourcing decisions in the SC profitability; model suitability for seasonal vs. non-seasonal products. The SC Modelling framework generic and globalising approach means that is easily applied and transposed to any other business realities and it can be easily changed to reflect other SC scenarios. The costing model associated means that, at any point in the network, all costs and profits can be easily measured. For the first time the shelf-life of a product captured and losses of product due to BBE dates, quantified. In this model the optimisation methodology runs parallel to the developed simulation tool, so the optimisation should be only run for new scenarios.