What hinders manufacturing companies from providing effective maintenance services? Researchers identified three key factors
Quality of the data, interoperability of different information systems, and data extraction. According to the findings of a recent doctoral dissertation, these are the key challenges that hinder manufacturing companies from carrying out effective, data-driven maintenance services.
Manik Madhikermi, M.Sc. (Tech.), will defend his dissertation on June 10 at Aalto University School of Science. He and his colleagues conducted research on Finnish multinational manufacturing companies. Using quantitative and descriptive analyses, they identified the major challenges these companies have.
For manufacturing companies, after-sales services like maintenance are as important as price and product features. They contribute to the client’s decision to buy, influencing the company’s sales, profits, and market share. Therefore, it is crucial to recognize what types of factors influence the capability to provide such services.
Firstly, researchers at Aalto University discovered the data that manufacturing companies use for decision-making are often of bad quality. This means that the data are useless for many segments of the company.
Secondly, companies tend to use different kinds of operational systems to streamline different processes, such as remote monitoring and maintenance management. Operational systems often come from different vendors, and according to the researchers, they do not communicate well with each other.
The third identified issue was data extraction. In practice, manufacturers tend to hide the data in their database. ‘That means that other people cannot reverse-engineer it,’ Madhikermi explains.
Solutions to real-world problems
In addition to identifying problems that manufacturing companies have, the research group developed methods and frameworks that can help to tackle those issues.
The researchers developed methods to assess and compare data quality of maintenance reports, an interoperability framework for multiple systems, and a data discovery methodology. The applicability and quality of the methods and frameworks were validated through one or multiple use cases.
The first question, that you need to have in mind is, how does this fit into my thesis?
For Madhikermi, the most interesting part of his research is its relevance to the real world. ‘My research is applied and it’s tackling problems that are real for the industry. The field is quite diverse. Some people are working in security and I am more focused on data engineering, especially Internet of Things in maintenance.’
Madhikermi finished his doctoral dissertation under the supervision of Professor Kary Främling.
Doctoral student, think what fits to your thesis
Rather than technical skills, Madhikermi thinks that the most important things he learned during his doctoral studies were organizing skills and seeing the holistic picture of a problem. He sees the ability to break a problem down into smaller pieces and rebuilding it again into the whole picture as very beneficial.
Madhikermi points out that one of the challenges of working on multiple projects is finding a way to put them all together. After his first year of doctoral studies, Madhikermi decided to focus strictly on projects that are applicable to his thesis.
This is his advice for prospective doctoral students: ‘Before writing any paper or doing any work, the first question, that you need to have in mind is, how does this fit into my thesis?’
Madhikermi’s background is in industry. After finishing his bachelor’s degree, he worked as a Senior Software Engineer for years. Madhikermi, originally from Nepal, first came to Finland because of his Master’s studies at the Tampere University of Technology.
Now the future is open, but Madhikermi’s plan is to combine research with work in industry. ‘For the next five years, I see myself working in the industry as a research scientist. I already got one job offer in my hometown in Nepal.’
Manik Madhikermi, M.Sc. (Tech.), will defend the dissertation “Moving towards data-driven decision-making in maintenance” on Monday 10 June 2019 at 12 noon in Aalto University School of Science, lecture hall TU2, Maarintie 8, Espoo.