Abstract
This work introduces a formation and variety of decision-making models based on operations research modeling and optimization techniques in smart manufacturing environments. Unlike traditional manufacturing, the goal of Smart manufacturing is to optimizing concept generation, production, and product transaction and enable flexibility in physical processes to address a dynamic, competitive and global supply chains by using intelligent computerized control, advanced information technology, smart manufacturing technologies and high levels of adaptability. While research in the broad area of smart manufacturing and its challenges in decision making encompasses a wide range of topics and methodologies, we believe this chapter provides a good snapshot of current quantitative modeling approaches, issues, and trends within the field. The chapter aims to provide insights into the system engineering design, emphasizing system requirements analysis and specification, the use of alternative analytical methods and how systems can be evaluated.
TopBackground
Mathematical models and optimization techniques are the driver for model-driven DSS. With regards to the structure of data and a problem’s objective and constraints, many programming tools and mathematical algorithms are available to aid decision-makers in building a DSS with optimal recommendations. The critical step is to know the type of optimization algorithm needed to solve the problem. For more details on a taxonomy of optimization problems, one can refer to a comprehensive collection of optimization resources at https://neos-guide.org/.
Mathematical algorithms support convergence towards optimal solutions. This review classifies optimization problems in terms of traditional and intelligent approaches. The most commonly used intelligent optimization models are search-based (i.e., metaheuristic models), learning-based (i.e., machine learning models), uncertainty-based (i.e., robust optimization; stochastic optimization), simulation-based (i.e., Markov Chain Monte Carlo) and Markov Decision Process (MDP) (see Tao et al., 2016, for a comprehensive review on intelligent optimization).
Although using an intelligent optimization algorithm can gradually adapt a specific model-driven DSS for smart manufacturing, such a DSS requires several other criteria be met to be adequately intelligent. More intelligent DSS are created with a learning algorithm, a knowledge sharing system, and with cognitive computing capabilities. Nevertheless, in a smart manufacturing system, with connectivity among all manufacturing processes, an intelligent, integrated DSS is required to manage a manufacturing system. Features of an integrated, intelligent DSS include expert knowledge, risk management, production control, quality monitoring, marketing and sales management, project management, and supply chain (SC) support. Guo (2016) provides an extensive collection of DSS capabilities and features needed for managerial tasks of smart manufacturing integrated with intelligent optimization algorithms.
Key Terms in this Chapter
Semantic Integration: The process of integrating information from diverse sources. In this regard, semantics focuses on the organization of, and action upon, information by acting as an intermediary between heterogeneous data sources which may conflict not only in structure but also in context or value.
Case-Based Reasoning: The process of learning to solve new problems based on the solutions of similar past problems.
Advanced Planning and Scheduling: Refers to a manufacturing management process by which raw materials and production capacity are optimally allocated to meet demand.
Markov Decision Process: Describes the environment for solving the optimization problem by reinforcement learning or dynamic programming. Provides a mathematical framework for modeling DM in situations where outcomes are fully or partially observable.
Order Penetration Point (OPP): Defines the stage in the manufacturing value chain, where a particular product is linked to a specific customer order.
Enterprise Resource Planning: The real-time integrated management of core business processes, mediated by software and technology.
Manufacturing Execution System: Computerized systems used in manufacturing to track and document the transformation of raw materials to finished goods.