Data analysis has become a crucial tool to support decision-making in many companies. The research project Data Analysis for Knowledge Intensive Product Development (DATAKIND) aims at developing new and improved algorithms and methods for data analysis to support the partner companies of the project with processes related to three areas of research: prediction with confidence, interpretable machine learning and real-time and distributed data analysis.
The goal of data analysis is to support decision-making by extracting useful information from data sets. Data analytics utilises generic algorithms and tools, capable of addressing a wide range of problems.
The industrial partners engaged in this project are Stena Line, an international transport- and travel service business with the most comprehensive route network in Europe, and ITAB, one of Europe’s leading suppliers of shop concepts, e.g., checkouts, lighting, fixtures, fittings and high-tech equipment. Both companies are digitally mature, in the sense that they have good access to high quality data, but they are at the same time in need of new and improved tools and techniques that can help them utilise the potential in their data when realizing knowledge intensive products.
Stena Line is undergoing a massive digital transformation and will become a cognitive company. Data analytics is a core component in their effort to realize improvements of the wide variety of processes within Stena Line. These include problems related to travel, hotel, retail and operational aspects.
ITAB is currently initializing research and development of solutions for a queue-free shopping experience, called AirFlow, which is aiming at a secure, simple and time-saving solution for both customers and shop owners. The task is complex and involves designing and building several new systems in which data analytics is required as a core functionality. The tasks involve, but are not limited to, customer tracking, detection of anomalous behaviour and optimization of sensor arrangement.
The project is organized around three areas of research highly relevant to the field of data analysis:
- Prediction with confidence
- Interpretable machine learning
- Real-time and distributed data analysis
Research and collaboration will be carried out using the Technical Action Research (TAR) approach, proposed by Weiringa and Morali (2012).
The purpose of this research project is to develop new and improved algorithms and methods to support the companies with processes related to the three areas of research. Using TAR, the project aims to identify problems common to both collaborating partners and based on these common problems develop generic solutions.
Co-production with both partners is organized around the software prototypes, which implement the artefacts being developed through a joint effort combining competences from both academia and partners, using the TAR methodology. The artefacts will include modifications of the conformal framework to new data representations, best practices regarding conformal predictors and Venn predictors, techniques for increased interpretability, edge-enabled prediction frameworks, as well as parallel and distributed predictive algorithms.
On top of that, co-authored or co-produced scientific publications will present the results to the scientific community. The goal is at least one co-produced publication per year from each work package. Furthermore, collaboration and joint prototyping across work packages will be encouraged when possible. However, each work package is responsible for meeting its objectives.
Both partners are expected to utilize machine learning extensively when developing future products and services, so working with and getting access to state-of-the-art methods and algorithms in this project, will also have a long-term positive effect.
- Stena Line
Project duration and financing
The project runs until January 2022 and is co-funded by the Knowledge Foundation.
If you would like to know more about the project, please contact Tuwe Löfström.
Content updated 2020-07-02