Adaptive iterated stochastic metaheuristic to optimize holes drilling path in manufacturing industry: The Adaptive-Dhouib-Matrix-3 (A-DM3). (April 2023)

• Record Type: Journal Article
• Authors: Dhouib, Souhail; Zouari, Alaeddine
• Is Part Of: Engineering applications of artificial intelligence. Volume 120(2023)
• Publication Date: 2023-04
• Subjects: Computer numerical control machine -- Holes drilling path -- Artificial intelligence -- Optimization -- Metaheuristic -- Dhouib-Matrix-TSP2 (DM-TSP2) -- Dhouib-Matrix-3 (DM3); Engineering -- Data processing -- Periodicals; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; Ingénierie -- Informatique -- Périodiques; Intelligence artificielle -- Périodiques; Systèmes experts (Informatique) -- Périodiques; Artificial intelligence; Engineering -- Data processing; Expert systems (Computer science); Periodicals; 620.00285
• DOI: 10.1016/j.engappai.2023.105898
• Languages: English
• ISSNs: 0952-1976

Abstract

Abstract: Tool path optimization approaches are often used to reduce production time and cost, energy consumption, etc. on Computer Numerical Control Machines. Thus, several artificial intelligence algorithms based on Traveling Salesman Problem (TSP) have been implemented to optimize tool trajectory length in various production sectors mainly the holes drilling process. In this field, this article exhibits first a related works survey detailing the main used algorithms. Then, it suggests a novel approach for optimizing non-productive tool paths for drilling a hole series. The proposed approach entitled Adaptive-Dhouib-Matrix-3 (A-DM3) is based on combining the iterated stochastic Dhouib-Matrix-3 (DM3) with a tabu memory inspired by Tabu Search (TS) metaheuristic. To confirm its ability and stability to find the shortest drilling tool path, A-DM3 method is tested on six practical case studies of a rectangular matrix of holes. Further, it is compared to some commonly used algorithms such as Genetic Algorithm (GA), Ant Colony Optimization (ACO) and some of its derivates, modified Shuffled Frog Leaping Algorithm (mSFLA), Cuckoo Search (CS), and hybrid Cuckoo Search Genetic Algorithm (CS-GA). Computational results proved the superiority of the proposed A-DM3 compared to these well-known metaheuristics in literature, particularly in a medium and large number of holes. Hence, A-DM3 provided a new shortest path length record with an improvement exceeding 100% in some cases compared to competing algorithms.