An expert system for the design of a multi-echelon inventory/distribution fulfilment system

In a multi-echelon logistic network every business actor generally tries to maximise the profit of its firms and not the profit of the whole chain in the interest of the final customer. For this reason the sharing of demand information among the entities of a logistic network reveals a possible solution to counter some of the distortions and fluctuations on inventory levels and products flows. Object of this study is the development of an innovative expert system to support the design and management of inventory/distribution logistic channels. Several performance indexes have been introduced in order to quantify and control the customer service levels and overall logistic costs. The proposed approach, which is based on the integration of different techniques (such as object-oriented simulation, what-if analysis and factorial analysis), offers the opportunity to reach a global system optimisation. A large set of results are obtained by a multi-scenarios analysis which identify and measure the most critical factors and combinations of factors affecting the performance collected along the whole system.