Production System Development: Enhancing Efficiency with Industry 4.0 Technologies

Production System Development (PSD)

(PSD) includes both the improvement of current production systems and the creation of new ones. The production process should be created in conjunction with the product, as representative of the entire design construction process, rather than after the brand has been already conceived (Bellgran and Säfsten, 2010). PSD examines the manufacturing operation and its conceptual design holistically throughout the life cycle of the system. Increasing production efficiency with PSD has long been considered one of the greatest methods to boost competitiveness. Change-ready production systems, which need good PSD, are becoming more popular as customer requirements and market possibilities change quickly (Andersen, 2020).PSD emphasizes modernising existing production systems while ultimately creating new ones. Material handling enhancements may vary from small incremental changes to completely new and ground-breaking ones.These have typically been optimised value-adding activities and many materials handling elements (Enke et al. 2018).

Advancements of production systems can be found in the concept of industry 4.0, that is a new frontier formanufacturing planning (Bueno et al., 2020; Cadavid et al., 2020). Industry 4.0 can be described as anintegration of information and communication systems with production facilities, supply chains and service systems within industries (Öner et al., 2018). These manufacturing technologies allow companies to developtheir production system, within aspects such as time and cost of a productbut also working towards the enabling of larger product varieties (Zheng et al. 2018). Not only are there enablers with the technologies but also several challenges when implementing new technology. A challenge that companies are confronted with is thevast amount of possibilities that come with the new technologies. Itis further stated that it can be hard for companies to choose the optimal technology for the company, in order to gain optimal results (Mourtzis, 2020).

A tool that can be used by companies to plan and evaluate risks at implementation of Industry 4.0 technologies is simulation (de Paula Ferreira et al., 2020). According to Ingemansson and Blomsjö (2004) Discrete Event Simulation (DES) is an important and powerful tool that can be used for designing and improving material flows, and evaluating different improvements. Supporting that claim, Mourtzis (2020) explains that Simulation technologies are a pillar of digital manufacturing solutions as it allows companies to experiment and validate different solutions, manufacturing configurations and processes without affecting the real system. The data thatis generated from the simulations enables companies to base difficult decisions on actual data and facts (Gajseket al., 2019). Several companies refrain from using simulation as they lack the required knowledge, as well as it being a time consuming attempt (Ingemansson and Bolmsjö, 2004).

Concept of Industry 4.0

Simulation can be used to create a virtual image of a system in a company in order to test functionality ofproposed changes, such as a conveyor belt system (Mikušová et al., 2019). Conveyor belts are one of the most vital systems of intra-company transportation in a wide range of industries (Fedorko, 2019). Indications on an emerging development of the conveyor belts have been detected due to advancements towards aligning with the concept of industry 4.0 (Liu et al., 2018). The idea ofsmart conveyors is emerging (Jurdziak et al., 2019),equipped with sensors and actuators, with the goal of collecting and understanding information to optimize operations (Soares et al., 2019). It is important to continuously research how various problem areas can betargeted, in order to facilitate reliable operations (Fedorko, 2019). Earlier research (Fedorko, 2019) has aimed atstudying whether industry 4.0 technologies can be used to solve issues connected to conveyor belts, in order to make operations more efficient. Another body of work (Fedorko et al., 2021) has researched the implementation of industry 4.0 technologies in the form of simulating a digital twin to suggest improvements to streamlineconveyor belts.

To boost organisational efficiency and capacity, organizations need to bring new production technologies or update current manufacturing methods (Moica et al., 2018). The key enablers to realize data-drivenmanufacturing are data collection and data analysis tools. Through these tools, physical models such assimulations to describe complex manufacturing processes, can be created. With the development of advancedsensors and data analyzing technologies, it is possible to build models with higher efficiency, accuracy and self diagnosis through extensive use of data. These models are still challenging to build; experiencing low accuracy or oversimplifications due to their complexity, they can create biases (Ke Xu et al., 2020). The vast amount ofpossibilities that is provided by the advancements of industry 4.0 technologies creates a challenge for companies to choose the optimal technology for the companies (Mourtzis, 2020). Although having the optimal technology for companies they tend to have issues related to disturbances and bottlenecks within their production systems (Abd Rahman, 2021).

The purpose of the study is to explore how simulation can be used to identify disturbances in a production system and which Industry 4.0 technologies that can increase productivity in a production flow. The purpose is summarized in the two following research questions:

RQ1: How can industry 4.0 technologies be used in order to streamline conveyor belt operations?

RQ2: How can simulation help identify disturbances in a production system?

Use of Simulation in Production Systems

The study is limited to researching a company in the manufacturing business, thus caution is required when generalizing results. The scope of the study is focused on a part of the full process at the production site whichis the conveyor system in the paint shop of the case company. The scope of the solution is therefore limited to a

specific area of the warehouse, which makes it not applicable to other companies or departments of the case company. However, through understanding how implementation of these technologies and workways can help, the study can work as a foundation for similar cases in the future.

The following segment presents the method that has been used during the process of the study. Initially, the research design and approach is presented, which establishes the chosen method of research. Thereafter, the method for the literature review and data collection is presented with a section discussing the reliability and validity of the data collection methods.

The study aims at providing a profound understanding of the current state of the production process and find what improvement measures can streamline production. To reach a proper insight into the operations of the company, a mixed research method was deemed fitting. The mixed method is a method that combines quantitative and qualitative research to take several perspectives into account (Bryman, 2012). The qualitative data is gathered through interviews and observations to get an understanding of the production site and the process layout. The qualitative data is also used to gather literature that can support the research to answer the research question. The quantitative data is gathered through time studies, and machine data that is used to build a simulation model that has been analyzed. Bryman (2012) further explains that the research method can be structured differently in a mixed method as there is no specific approach to a mixed method. In this case, an inductive approach is used based on the research reasoning up to a theoretical conclusion through data collection and analysis (Bryman, 2012). In this study, the inductive approach was used to understand how the different industry 4.0 technologies and new workways could benefit the existing production process.

A project introduction and smaller problem formulation of the case company was given by a university lecturer at MDH to give the project group an introduction to the case company and the project. To get a betterunderstanding of the production site and existing systems, a visit was conducted to get a better insight of the process. A conceptual model was built based on empirical data from the first visit and observations of the process to map down each process step, the existing technologies and the issues presented by the case company. Thereafter, a literature study was done to provide solutions for the research question on systems or new technology that could be implemented in the process steps. A second visit was made to conduct a semi-structured interview afterwards to gather further empirical data in relation to the theory gathered, and tobuild a simulation model. The empirical data was then compared to the theoretical background with theintention of analysing whether theory was applicable in reality. A simulation was built out of the new empirical numeric data to study, experiment and analyse the behaviour of the production system. The simulation creates a better understanding of the current state of the process which was evaluated. An analysis was then presented with possible solutions and a guide to what changes could be implemented. Lastly, a discussion was made of the findings and possible drawbacks with an ending conclusion to summarize the findings.

Smart Conveyor Systems and Industry 4.0

The data for the study has been collected through a literature study from research and empirical data that has been collected through interviews and observations at the case company. A purposive sampling method has been conducted to gather relevant data in the literature study, which according to Bryman (2012) means that literature is gathered with the purpose of the study as reference. Keywords such as “Automation”, “Conveyor”, “Industry 4.0”, “Production system development”, “Simulation” and “Smart manufacturing” were used and combined to search for information. Search engines or databases such as Google Scholar, IEEE Xplore, Scopus and Primo were used to find relevant sources of data. The literature sampled were scientific sources gathered from books,articles, journals or other types of scientific publications. Due to the large scope of different industry 4.0 technologies and possible solutions, only a few technologies were analyzed. A backwards snowball sampling method was used, where sources are used to reference new sources, with regard to the purposive sampling method (Bryman, 2012). Thus, this method was used to find new sources through the references of other articles and journals. The literature study is built on secondary data which is data that has been used for another purpose than this study (Bryman, 2012).

A main part of collecting empirical data was done through conducting interviews, which according to Säfsten and Gustavsson (2020) is a suitable method to gather data regarding a specific phenomena. With the purpose of this study, an interview has been conducted to gather valuable data from the case company to understand the conveyor belt and the existing system. A semi-structured interview was conducted, which is an interview with predetermined questions or discussion topics but allows room for unprepared questions (Bryman, 2012; Säfstenand Gustavsson 2020). The semi-structured interview was conducted with a Process technician, Maintenanceengineer and Production technician who have experience of the case company and the production site.

A group interview method was used when conducting the semi-structured interview. A group interview is an interview form where several people discuss a number of topics. This method allows the interviewees to add their opinions and experiences together and gives the researcher the ability to study how individuals collectively construct meaning of things. Furthermore, it allows the interviewees to create a discussion and add information that would not have been shared otherwise (Bryman, 2012). These respondents all have different positions, with different work duties, skills and personal experience which may influence the results. The group interviewb method was used to gather the perspectives and thoughts of the respondents. The respondents have been anonymized according to Table 1 in appendix.

The empirical study was also based on observations. A visit to the production site was done at the time of the first visit to understand the existing system and process. The observations were made to build a simulationmodel as well as to understand the layout of the stations. Furthermore, time studies were made of manual workto understand the average time it takes to finish some steps of the process depending on the product mix. Data that was available from other process steps, such as automated jobs, have also been collected and observed. These observations validate the information used for the simulation model so it is built correctly.

In order to establish the reliability of the study, it is important to keep an objective view on facts (Bryman and Bell, 2011), which has been followed throughout the process. Holme and Solvang (1997) present that a qualitative approach brings the advantage through the proximity created between the one studying the object and what is being studied. To further establish the reliability and validity of the gathered theory, only peer-reviewed articles have been used. Furthermore, verifying thepublication date, publisher, authors and their background has been done to confirm the reliability of the source. Another method used to validate sources has been to include several sources with similar information in certain cases. This further validates the reliability of data gathered.

For the empirical data, the group has interviewed three workers with years of experience at the production site together. The interviewees could validate or challenge each other's answers through a group interview which made sure that data gathered and information discussed was confirmed by all interviewees (Bryman, 2012). The observations that were made, have been conducted at the production site. The time studies conducted have been repeated several times to validate the reliability of the result. This ensures that the empirical data and data used for the simulation model is validated, and the analysis for the real world case is as true as possible.