Cell formation with operational time using ART1 networks

Cell formation problems are typically combinatorial optimisation problems and pose difficulties to obtaining quality solutions. Researchers have proposed various algorithms based on different approaches to obtain disjoint machine cells. The major limitations of these approaches lie in the fact that real-life production factors, such as operational times, lot sizes and sequence of operations for different parts are not taken into account. In the present work, an attempt has been made to propose an Adaptive Resonance Theory 1 (ART1) algorithm to handle the real valued workload matrix. ART1 algorithm is one of the types of Artificial Neural Networks that is used in many applications such as image processing, data clustering, pattern recognition, etc. It is one of the prominent approaches found in literature for cell formation problems. A Modified Grouping Efficiency (MGE) is proposed to measure the performance of the algorithm. The performance of the proposed algorithm is compared with that of the K-means method and Genetic Algorithm (GA). The results distinctly indicate that the proposed algorithm is quite flexible, fast and efficient in computation for cell formation problems and can be conveniently applied in industries.