Track: Production Planning and Control

Abstract

The current paradigm shift in manufacturing, widely known as Industry 4.0, exists at the nexus of advances happening in computer science, sensing technologies, Information and Communication Technology, and big data analytics. It brings together Internet of Things, social networking, and advanced analytics to meet the growing need of personalized production at lower costs, by integrating human-like social capabilities into the assets of an industrial system. In this paper, we propose an innovative Multi-Agent System based distributed operations planning approach for scheduling of jobs in a parallel machine shop-floor. The approach harnesses the capabilities of Cyber-Physical Systems formed by bringing together physical machines, and various functional divisions, with their cyber space, or agents. These agents interact with one another to form a network of social machines. Using distributed decision-making and communications within the network of social assets, we tackle the complex, NP-hard problem of job scheduling, and compare the results with that of conventional centralized operations planning approach. The advantages of the proposed approach are clear in terms of reduction in computation time and lateness, and the flexibility offered by the distributed approach.The current paradigm shift in manufacturing, widely known as Industry 4.0, exists at the nexus of advances happening in computer science, sensing technologies, Information and Communication Technology, and big data analytics. It brings together Internet of Things, social networking, and advanced analytics to meet the growing need of personalized production at lower costs, by integrating human-like social capabilities into the assets of an industrial system. In this paper, we propose an innovative Multi-Agent System based distributed operations planning approach for scheduling of jobs in a parallel machine shop-floor. The approach harnesses the capabilities of Cyber-Physical Systems formed by bringing together physical machines, and various functional divisions, with their cyber space, or agents. These agents interact with one another to form a network of social machines. Using distributed decision-making and communications within the network of social assets, we tackle the complex, NP-hard problem of job scheduling, and compare the results with that of conventional centralized operations planning approach. The advantages of the proposed approach are clear in terms of reduction in computation time and lateness, and the flexibility offered by the distributed approach.