114 A CPS-Agent self-adaptive quality control platform for industry 4.0 Narjes Alaya * ,Baudouin Dafflon § , Nejib Moalla * and Yacine Ouzrout * * University Lumire Lyon 2 § University Claude Bernard Lyon 1 Abstract— With the emergence of new requirements in man- ufacturing, mainly the product customization trend, imple- menting a solution for product quality control becomes an increasingly complicated task. In this modern industrial context, those solutions need to take into account the constant updates in the product design and combine its resources in the most effective manner in order to meet the new product specifica- tions. By reason of the lack of available technical solutions, this task is usually carried out by human operators which is time consuming, tiring and not always a reliable solution. In the present work, a self-* automatic quality control platform based on cyber-physical systems and multi-agents paradigm is presented. The solution is validated within a real industrial use- case. Keywords: Industry 4.0, Cyber-Physical System (CPS), Multi- Agent System (MAS), Automatic Quality Control (AQC). I. INTRODUCTION The modern industrial context is characterized by two im- portant dimensions: On one hand, the need for collaboration in the form of companies-networks in order to face competi- tiveness especially for small and medium enterprises. On the other hand, the need of enabling products customization to better meet customer needs and exigences. Traditional manufacturing approaches have proven unable to address the emergent needs mainly because of their centralized and rigid conception. Consequently, nearly all of the recent production paradigms seek to adopt more agile approaches enabling responsiveness, robustness, self- *, and flexibility. These new needs have coincided with huge advancements in the technological area, mainly high availability and affordability of sensors, computer networks, the Internet of Things and the Artificial Intelligence.These circumstances have been at the origin of the industry 4.0 paradigm emergence. This new concept, which was used for the first time in 2011 in the Hannover Fair in Ger- many, involves the main technological innovations applied to production processes in the field of automation, control and information technologies [6]. It enables companies to schedule maintenance, predict failures and adapt themselves to new requirements and unplanned changes in the produc- tion processes in an autonomous way [7]. The fourth industrial revolution led to introducing the Cyber Physical Systems (CPS) as a new technological enabler allowing to build modular, intelligent and flexible systems. According to Lee [10], CPS are defined as autonomous systems, have a self-* behaviour, and are able to make decisions; this intuitively leads to consider Multi-agents system paradigm [24] as a valuable alternative to design CPS. Actually, agents are able to respond promptly and correctly to changes, due to their inherent capabilities of self-adapting to emergent events without external intervention [12]. In the present work, we are interested in automatizing the product quality control which represents one of the key application domains of the industry 4.0 for several reasons: • It enables an efficient collaboration since it guarantees a better conformity to the product specifications. • It allows partners to track the product state during the different manufacturing phases and have an idea about eventual problems. • It gives the customers the possibility to update on-the- fly the specifications of the product to be manufactured without interrupting the process. . Traditionally, quality control is performed according to a test plan where the measurements and the decision criteria are defined. Those specifications are usually updated within a human intervention, which is suitable for mass production. But with today’s growing need for product customisation, the test plan should be constantly updated and configured because of the huge variety of products. This task is no longer possible to be performed by humans. For this reason, proposing an autonomous self-* quality control approach becomes an essential task. In this paper, a CPS-Agent based quality control solution, in line with the industry 4.0 paradigm, is proposed. At this early development phase, there is a need to clearly define the CPS conception. In the following section of this paper, a proposal of a CPS design inspired by multi-agents systems is introduced. The proposed architecture is then used to implement a CPS-agent based framework for quality control. The framework will be finally validated within a real industrial use case. II. THEORETICAL BACKGROUND Advances in technology have engendered the Internet of Things (IoT) emergence [1] where embedded sensors are in charge of collecting data and sharing it through the network. Today’s technology is unable to store and process the huge amount of data generated through this process, this phenomenon is known as the Big Data challenge. In order to address this problem, many innovative concepts have been proposed particularly in the industry 4.0 context. One of the most interesting key-enablers that have drawn attention in the manufacturing field is the Cyber Physical System which is considered as an autonomous and reactive entity that interacts with its physical and logical environment. It 7th International Conference on Information Society and Technology ICIST 2017