ART based cell formation using combined operation sequence and time

The cell formation (CF) problem mainly deals with clustering of parts into part families and the machines into machine cells. The parts are grouped into part families based on similarities in their manufacturing and design attributes and the machines are allocated into machine cells to produce the identified part families. The zero-one part-machine incidence matrix is commonly used as input to any clustering algorithm. The output is generated in the form of block diagonal structure. Production data such as operation time, sequence of operations, batch size etc. that have significant bearing on smooth flow of materials are not considered in such methods. In this paper, an attempt has been made to develop an algorithm based on adaptive resonance theory (ART) neural network to addresses this issue by considering combination of operation sequence and operation time of the parts to enhance the quality of the solution obtained for the CF problem. A new performance measure is proposed to assess the goodness of the solution quality obtained through proposed algorithm. The performance of the proposed algorithm is tested with example problems and the results are compared with the existing methods found in the literature. The results presented clearly shows that the performance of the proposed algorithm is comparable with other methods for small size problems and better for large size problems.