Towards IoT-enabled Dynamic Service Optimal Selection in Multiple Manufacturing Clouds
Chen Yang,Tao Peng,Shulin Lan,Weiming Shen,Lihui Wang
DOI: https://doi.org/10.1016/j.jmsy.2020.06.004
IF: 12.1
2020-01-01
Journal of Manufacturing Systems
Abstract:With the Internet of Things, it is now possible to sense the real-time status of manufacturing objects and processes. For complex Service Selection (SS) in Cloud Manufacturing, real-time information can be utilized to deal with uncertainties emerging during task execution. Moreover, in the face of diversified demands, multiple manufacturing clouds (MCs) can provide a much wider range of choices of services with their real-time status. However, most researchers have neglected the superiority of multiple MCs and failed to make a study of how to utilize the abundant and diverse resources of multiple MCs, let alone the multi-MCs service mode under dynamic environment. Therefore, we first propose a new dynamic SS paradigm that can leverage the abundant services from multiple MCs, real-time sensing ability of the Internet of Things (IoT) and big data analytics technology for knowledge and insights. In this way, providing optimal manufacturing services (with high QoS) for customers can be guaranteed under dynamic environments. In addition, considering that a relatively long time might be spent to complete a complex manufacturing task after SS, a quantified approach, based on the Analytic Hierarchy Process and big data, is proposed to evaluate whether the intended cloud manufacturing services should be reserved to make sure that eligible services are ready to use without compromising cost or time. In this paper, the problem of IoT-enabled dynamic SS across multiple MCs is formulated in detail to enable an event-driven adaptive scheduling when the model is faced with three kinds of uncertainties (of the service market, service execution and the user side respectively). Experiments with different settings are also performed, which show the advantages of our proposed paradigm and optimization model.