Towards the Future Architecture of Smart Manufacturing Systems

Sammanfattning: Background: Automotive industries have started to digitalize their manufacturing processes by developing new architectures that unite from the early stages of development. This study finds the Requirements Engineering (RE) practices and the challenges organizations face for new architectural developments. We have found that over 57% of the practitioners do not revise their initial requirements until these requirements are satisfied. This can cost a lot of time and money at later stages of development. The RE practices are established well in industries but, organizations (Volvo group) may vary on how these practices are integrated and lead to different outcomes. This study tries to find the techniques used and the challenges faced by the organizations in new developments by analyzing existing architectures and constructing a Software Requirement Specification (SRS) document for a new architecture at Volvo group. Objectives: This thesis aims to comprehend the significance of RE techniques and the difficulties experienced in conjunction with new architectural advances in the automotive industry.Methods: The methods used in this study are a case study, a survey, and a literature study. The data obtained from these methods are utilized to create a survey with open questions. The case study was used to understand the current architectures and the survey was used to obtain data as it helps to reach a larger audience. The literature study was conducted with all the keywords obtained from the survey and interviews to find the real-world impact of the challenges. Results: From the interviews, we found new architectural developments have no explicit roles. A single developer may act as a Requirements Engineer and tester until their requirements are certain. Techniques used to elicit may differ from the organization. Most of the participants from the survey have chosen Workshops and Brainstorming. The survey has recorded 37 responses from participants of diverse domains with unusual inputs about the challenges faced in new developments. We believe that our analysis provides detail about challenges faced by the industry and helps in understanding the process of new architectural developments. Conclusions: RE methods vary according to the type of product developed. So, the methods of development change according to their domain. New developments often lead to unknown territories and organizations may require a specific custom method to pursue a new architecture development. We develop an architectural design by following the RE principles and techniques in the automotive industry. We generalize our findings to new developments by utilizing surveys and literature studies. We conclude our research by providing strings obtained from our study that show the real-world impact of the development in the future of smart manufacturing.