A framework for smart production-logistics systems based on CPS and industrial IoT
Abstract:
Industrial Internet of Things (IIoT) has received increasing attention from both academia and industry. However, several challenges including excessively long waiting time and a serious waste of energy still exist in the IIoT-based integration between production and logistics in job shops. To address these challenges, a framework depicting the mechanism and methodology of smart production-logistics systems is proposed to implement intelligent modeling of key manufacturing resources and investigate self-organizing configuration mechanisms. A data-driven model based on analytical target cascading is developed to implement the self-organizing configuration. A case study based on a Chinese engine manufacturer is presented to validate the feasibility and evaluate the performance of the proposed framework and the developed method. The results show that the manufacturing time and the energy consumption are reduced and the computing time is reasonable. The presented work potentially enables manufacturers to deploy IIoT-based applications and improve the efficiency of production-logistics systems.
INTRODUCTION:
The increasing demands for customized products and services as well as frequent market fluctuations have posed challenges to management and control of manufacturing processes. For example, due to external and internal fluctuations, such as production order changes and unexpected equipment failures, production plans and schedules become inefficient or even infeasible. In addition, customized production with a small batch and short production cycle increases the computational complexity and requires more time cost for scheduling. As a consequence of lacking collaboration, excessively long waiting time and large amounts of energy are wasted in production and logistics. The emergence of advanced technologies, such as Internet of Things (IoT) [1], cloud manufacturing (CMfg) [2], cyber-physical systems (CPS) [3]–[5], and service-oriented technology (SOT) [6], has provided several promising opportunities to address the aforementioned challenges. The rapid development and widespread use of industrial IoT (IIoT) technology in manufacturing industry greatly promote information progress in real-time monitoring, traceability, tracking, transparency, and interaction [7]. Real-time and multi-source manufacturing data generated by embedded devices and sensors have been used to perform operation optimization and decision-making [8]. CPS with integrated computational and physical capabilities has been used to implement the efficient management and utilization of big data [9]. Besides, a variety of IIoT-based models and applications are developed to improve the efficiency of manufacturing industry [10]. With respect to the topic of production and logistics, many researchers and practitioners focus on the simultaneous scheduling of machines and automated guided vehicles (AGVs) [11], [12]. Although a few of them paid attention to the IIoT-based synchronized relationships between production and logistics, which have shown improvements in overall performance of enterprises operations [13]–[15]. However, existing manufacturing paradigms are insufficient to address typical problems of production logistics in job shops. These problems are listed as follows.
(1) For manufacturing resources in the infrastructure layer of job shops, how to achieve manufacturing status perception and intelligent modeling on the key manufacturing resources side, such as machines and T 3 material handling systems?
(2) For manufacturing tasks in the job shop level, how to implement a task-driven smart manufacturing service chain to realize active response and optimization of global interaction and collaboration?
(3) For executive processes of production logistics in job shops, how to conduct real-time performance analysis, exception diagnosis, and self-adaptive conflict resolution to realize real-time interaction and selforganizing configuration between machines, materials, and human?
Here, the authors present a framework for smart production-logistics systems (SPLS) and investigate the mechanism and methodology of SPLS. The investigation is focused on two key problems, namely the intelligent modeling of manufacturing resources in the infrastructure layer and the self-organizing configuration of smart manufacturing service groups. The proposed conceptual framework of SPLS is validated by a case study based on a Chinese engine manufacturer, showing better potentiality than the separated production logistics at each level of manufacturing services. Based on the developed engine manufacturer framework, a comparison is conducted to show the key features of the separated production logistics and SPLS. In Fig. 1, four levels are involved, including equipment level, job shop level, enterprise level, and industry level.
At the equipment level, for the separated production logistics, machines and material handling systems are managed and controlled in a passive manner, i.e., production tasks are usually assigned to machines. In contrast, SPLS makes machines and material handling systems ‘smart’ by using CPS and IIoT technology, which are capable of active perception, active response, and autonomous decision-making. For example, 5 machines actively request production tasks instead of waiting for an assignment. At the job shop level, each individual production and logistics is scheduled separately. To overcome the inefficiency and infeasibility in the scheduling which arises from both external and internal dynamic changes, production and logistics are integrated into a smart control system such that SPLS is capable of exception identification, self-organizing configuration, and self-adaptive collaboration.
At the enterprise level, multi-source and heterogeneous data collected from separated production logistics might require a considerable amount of computational resources to support decision making. Different from separated production logistics, capabilities of manufacturing resources for SPLS are encapsulated into smart manufacturing services using cloud computing technology as well as real-time and multi-source data. Manufacturing services are published on the cloud platform to complete production tasks in an on-demand manner. At the industry level, most enterprises are independent, whose production information is not shared. As a consequence, separated production logistics might waste manufacturing resources due to a low rate of usage, which causes more manufacturing costs, manufacturing time, and energy consumption. Benefiting from the high degree of integration between production and logistics, SPLS can implement the self-organizing configuration of manufacturing resources not only within a job shop but also between cooperative enterprises, which may increase resource utilization, eliminate waste, and improve productivity in the manufacturing industry.
To summarize, the objectives of this research work are the following: (1) to investigate the mechanism of intelligent modeling and active response of manufacturing resources in the infrastructure layer; (2) to investigate the mechanism of self-organizing configuration for SPLS. The remainder of this paper is structured in the following way. A review of the existing literature relevant to this study is included in Section II. The overall architecture of SPLS is described in detail in Section III. Section IV explains the intelligent modeling of key manufacturing resources based on CPS and IIoT. The mechanism and methodology of SPLS are presented in Section V.
CONCLUSION:
In order to cope with frequent changes and disturbances, discrete manufacturing systems require a high level of integration between production and logistics. This paper introduces a conceptual framework of SPLS and the mechanism and methodology of self-organizing configuration for collaborative production-logistics. Two problems in the field of manufacturing are addressed, including the intelligent modeling of manufacturing resources in the infrastructure layer and the self-organizing configuration of smart manufacturing service groups. The research is carried out to achieve the self-organizing configuration of SPLS based on CPS and IIoT. In the proposed SPLS, manufacturing resources at all levels are capable of responding to disturbances actively and coordinating intelligently. Bi-directional interaction of production-logistics and collaborative relationships between machines, materials, and human are achieved based on the proposed self-organizing configuration mechanism. As a consequence, production-logistics systems can be optimized adaptively and collaboratively when exceptions occur.