Research Information System
22nd International Symposium for Production Research, ISPR 2022, Antalya, Turkey, 6 - 08 October 2022, pp.770-779, (Full Text)
This study solves a one-dimensional cutting stock problem with multiple stock lengths. It is applied in a manufacturing setting where rolls of steel rods of different lengths are cut according to customer requirements. The one-dimensional cutting stock problem (CSP) is an NP-hard problem, including discrete demands and capacitated planning objectives. It is solved using column generation techniques. This study aims to develop a production plan that minimizes the waste of cutting steel rods of different lengths and diameters in required lengths. The approach to solving the problem has two steps. The first step is a heuristic algorithm that produces a cutting pattern at every iteration, which is then fed into a novel mathematical model to determine an optimal solution. An initial solution is obtained using randomly generated cutting patterns for the mathematical model. The algorithm terminates after a given number of iterations. The paper also proposes a Decision Support System, addresses application issues, and concludes with further studies.