Knowledge-based simulation for supply chain integration

Abstract

In a competitive business environment, it is necessary to adopt a new supply chain strategy from time to time in order to increase competitiveness. However, most small and medium-sized enterprises (SMEs) do not have the guidelines necessary for selecting the most suitable strategy to quickly achieve success. If they use the traditional trial and error approach, a great amount of effort and heavy costs would be incurred. Once an unsuccessful strategy is implemented, the loss to the parties involved would be considerable. In supply chain management, demand uncertainty is the most critical problem that most parties have to face in their current practices. To cope with the problem, retailers attempt to keep a high level of inventory or order larger quantities than needed from their suppliers. The high inventory costs are charged to the retailers. A vendor managed inventory (VMI) strategy is a way of solving such kinds of problems. Low inventory levels will be kept by the retailers and the lead time will be reduced. Fast responsiveness to customers is one of outcomes that can be achieved by the VMI strategy. It is also a key factor to drive the supply chain to success. However, the adoption of a new supply chain strategy needs a large amount of effort. Besides, most SMEs lack a simulation platform to mimic the effect of a change of supply chain strategies. It would be better if the SMEs predict the effects, caused by a new strategy, at the redesigning stage. This will help with reduction in the level of inventory, lead time, unfilled rate and stock-out rates. The platform can thus play an important role in the selection of a suitable supply chain strategy, at the preliminary stage. In the present study, a knowledge-based simulation platform (KBSP) is proposed to enhance the competitiveness of all parties in the supply chain. It aims to accumulate successful experiences that SMEs have had in formulating and implementing competitive supply chain strategies. Based on many unpredictable factors, the KBSP is provided for multiple retailers and a supplier who can simulate the processes of a VMI strategy. The outcomes of the implementation of the supply chain strategy can be derived from the simulation. The potential benefit of the KBSP is to provide a simulated environment for the SMEs, which will reduce the bullwhip effect due to demand uncertainty. A KBSP prototype has been designed and developed, and a trial run for evaluation of its performance has been conducted at a selected reference site. During the simulation, spectrum analysis is used to transfer patterns of demand to a series of spectra from which features of the patterns are extracted for comparison. History logs of processes in the simulation and terms in an agreement are valuable knowledge. This knowledge is captured, reused and retained in a knowledge repository (KR). The knowledge contains, for instance, problems that are solved, and details of how and why certain decisions were made. New staff are able to learn about a number of successful cases which have been accumulated in the KR. Angus Electronics Company Ltd. was selected as the site for the case study in which the system was trial implemented. In this case, the VMI strategy was selected for the simulation of the trial implementation. Compared with the traditional approach, that is, the push system, the simulation results show how inventories and the stock-out rate were significantly reduced, and how the fill rate was increased. The change of the strategy from the push system to the VMI strategy indicates that the strategic change is beneficial for Angus in that it can reduce the level of inventory and reduce the lead time. The theoretical background and know-how of this study forms the basis for further development of the proposed KBSP to handle complex supply chain problems. It is suggested that KBSP should be further developed and implemented in manufacturing company. Once the KBSP is further implemented successfully, the risks, such as demand uncertainties, would be identified and reduced. In addition, it is proposed that the KBSP is integrated with enterprise resources planning (ERP) systems, which provide a more realistic environment for different parties to conduct simulations. It is suggested that the KBSP is further developed so that more supply chain strategies can be made available for parties who plan to redesign their current strategy.