No abstract

In today's time-sensitive markets, effective storage policies are widely accepted as a means for improving the efficiency of order picking. As a result of customization, the variety of products handled by a warehouse has increased, making storage location assignment problems more complicated. Different approaches have been proposed by researchers for improving storage assignment and order picking. However, many industrial practitioners find it difficult to adopt such approaches due to complexity and high associated costs. In particular, small and medium enterprises (SMEs), that generally, lack resources and who have staff members with weak artificial intelligence backgrounds, still rely on experience when assigning storage locations for diverse products. In these circumstances, the quality of decision making cannot be guaranteed. In view of this, an intelligent system which can be easily adopted by SMEs is designed to improve storage location assignment problems. The proposed system, an RFID-based storage assignment system (RFID-SAS), is a rule-based system incorporating radio frequency identification (RFID) provides decision support for storage assignment in a warehouse. Unlike many existing situations, RFID tags are attached to products at the item level instead of at the pallet level. As the knowledge embedded in the system is represented in the form of rules, evaluation is important and is outlined in this paper. The effectiveness of the system is verified by means of a case study in which the system is implemented in a typical SME specializing in machinery manufacturing. The results illustrate that RFID-SAS can enhance the efficiency of order picking in a warehouse.

No abstract

No abstract

In this paper, a supervised self-organisation Neural Network (NN) for direct shape from shading is developed. The structure of the NN for the inclined light source model is derived based on the maximum uphill direct shape from shading approach. The major advantage of the NN model presented is the parallel learning or weight evolution for the direct shading. Here the proved convergent learning rule, the rate of convergence and a zero initialisation condition are shown. To increase the rate of convergence, the momentum factor is introduced. Furthermore, the application of the network on IC (Integrated Circuit) component shape reconstruction is presented.