Process cost prediction: a soft computing approach

Cost estimation based on expert's judgment is not an ideal approach, since human decisions are usually determined according to general attributes of limited and unstructured experience. The purpose of this paper is to develop a generic model of intelligence and cognitive science‐based method that can play an active role in process cost prediction within the shortest possible time.

In this paper, an intelligent system was conceived for prediction of total process cost of the product. The system is based on the concept of case‐based reasoning. It is a method for solving problems by making use of previous (source cases), similar situations and reusing information and knowledge about such situations. The source case data are generated by Taguchi technique and the cost function calculates the corresponding cost of each experiment in the economic time scale. The target case consists of the process variables whose cost needs to be determined. The cost for the source cases, consisting of the process variables of the already manufactured products are known in priori. The system calculates the similarities between the source cases and target cases and calculates the optimum cost. The fuzzy‐C‐means clustering method provides the model connecting the process parameters with total costs searched for.

The results show that the quality of predictions made by the intelligent system is comparable to the quality assured by the experienced expert. The proposed expert system is superior to traditional cost accounting system and assists inexperienced users in predicting the optimum process cost within the shortest possible time.

The research was limited to the traditional machining process.

The paper can be applied to any process industry and will have immense practical value.

This is the first time an expert system has been developed for the process industry that can calculate the process cost within a few days or a few hours before making an offer to a buyer.