Research Projects

Current Projects

Currently, our group has the following research projects.

• Research Council of Finland (former Academy of Finland) funds the project DESIDES (Decision Support with Interactive Advanced Data-Enabled Multiobjective Optimization Systems) in 2023-2027. In this project, we develop advanced, interactive multiobjective optimization methods to support making trustworthy and intelligible decisions. We also extend our focus to group decision making. The methods to be developed will augment data with human judgment and communicate decisions and their consequences to the people involved with novel, interactive visualizations. At the same time, we extend the open-source software framework DESDEO group decision making and answer the need to provide explainable, data-enabled decision recommendations. We will pilot and verify our approaches with selected application cases.
• Research Council of Finland (former Academy of Finland) funds the consortium project UTOPIA (Is climate smart forestry a utopia if the preferences of landowners are not considered?) in 2023-2025. The other consortium partners are Natural Resources Institute Finland (PI) and University of Eastern Finland. In this project, we develop methods to support stakeholders in implementing climate smart forestry, which advances carbon neutrality but also considers financial objectives. Since different forest owners typically have different preferences about what kind of management is practiced, we focus on supporting decision making in this setting and finding the most preferred solution among multiple conflicting objective functions.
• Research Council of Finland (former Academy of Finland) funds the Anti-fragile Decision-making: tHink beYonD Robustness And resilience (HYDRA) project in 2024-2028. HYDRA aims to provide support to make better and antifragile decisions. The focus is on handling deep uncertainty and better preparation for the future. The architecture to be developed renews the perspective of decision-making under uncertainty to think beyond robustness and resilience and look for opportunities to benefit from volatilities. This comprehensive architecture does not only rely on history and unreliable predictions. It also considers less probable and rare events and consequences of the decisions, supports simultaneous consideration of multiple objectives, provides adaptive pathways to avoid fragility and continuous performance improvement in the long-term planning horizon, and will be validated in real-life applications.
• Research Council of Finland (former Academy of Finland) funds the project Decision support for harvest scheduling under uncertain road accessibility in 2024-2025. The goal of this project is to develop the foundations for decision support where the uncertain availability of roads is taken into account already in the harvest planning and thus to develop a method that makes it easier for logistics planners to develop schedules for harvesting and transport that match both the demand and the road accessibility. In the long run, this is expected to enable forestry to create robust harvesting plans with greater accuracy and less risk of replanning than today, thus leading to better utilization of timber as well as planning, harvesting, and transport resources.
• We also have some confidential projects.

Past Projects

• The Peter Wallenberg Foundation funded a project Value of Information in harvest planning of Nordic forests in 2022-2024. Our consortium partners were Skogfors in Sweden and Department of Mathematics and Statistics at our university.
• Tietoevry funded a project in optimized 3D truck loading in 2023-2024.
• Academy of Finland funded the project DAEMON (Data-driven Decision Support with Multiobjective Optimization) in 2019-2023. In this project, methods were developed for data-driven decision support. Decision analytics was augmented with multiobjective optimization as multiobjective decision analytics. The results were implemented in an open source software framework DESDEO and are, thus, applicable for researchers and society. The project considered data and decision problems from selected fields of life to demonstrate the large application potential of the framework developed.
• Academy of Finland funded the thematic research area (profiling area) in the project DEMO, Decision Analytics Utilizing Causal Models and Multiobjective Optimization in 2017-2021. Even though the funding from the Academy has ended, DEMO still exists. DEMO focuses on explicit, concrete decision problems that can be presented with mathematical formalism. Predictive analytics, statistical modelling, causal inference, prescriptive analytics and multiobjective optimization are the key elements needed to create a seamless chain from data to decision. We refer to this as decision analytics. Thanks to method and software development, decision analytics is applied to support e.g. other profiling areas of the University of Jyvaskyla, especially related to education and health in dealing with their decision problems.
• Academy of Finland funded the project DESDEO (Decision Support for Computationally Demanding Optimization Problems) in 2015-2019. In this project, new interactive multiobjective optimization methods were developed for dealing with computationally demanding problems (e.g. when function evaluations are time-consuming or dimensions are high). In hybrid methods developed, expertise and strengths of different fields were combined. In addition. a novel, general interactive multiobjective optimization framework was developed. It includes open interfaces to connect external modules to it. With this project, interactive multiobjective optimization were brought closer to users and awareness of its potential was increased.
• In the FiDiPro project DeCoMo (Decision Support for Complex Multiobjective Optimization Problems), FiDiPro Professor Yaochu Jin, Chair in Computational Intelligence, Department of Computing, University of Surrey, UK worked in Jyvaskyla in 2015-2017. The project was funded by Tekes, Outotec, Fingrid. It involves also FIMECC, Fortum Power and Heat, Simosol, Valmet Power, Valtra and Honda Research Institute Europe.

  FiDiPro Professor Yaochu Jin is a world-leading researcher in surrogate-assisted evolutionary optimization as well as multiobjective optimization. He has rich expertise in learning systems, in particular in integrating evolution and learning. Additionally, he has long experience working with the Honda Research Institute, where he worked on various real-world aerodynamic optimization problems.

  In DeCoMo, the expertise of Prof. Jin and that of the Industrial Optimization group in interactive multiobjective optimization complemented each other and together we developed novel optimization methods for decision support in solving complex multiobjective optimization problems by combining modern meta-heuristics and machine learning techniques. The output of this project was a prototype of an intelligent decision support tool that can make advanced multiobjective optimization methods available for industry, thereby significantly enhancing the innovation capability and competitiveness of the Finnish industries.

• FINNOPT was a one year project (August 2014-July 2015) funded by Tekes. The aim of FINNOPT was to commercialize the state-of-the-art methods in optimization, developed at the Industrial Optimization Group.
• Tekes-funded project SIMPRO, Computational methods in mechanical engineering product development, 2012-2015, (joint project with VTT, Altoo University, Tampere University of Technology and Lappeenranta University of Technology) focused on methods and systems of high-performance computing in mechanical engineering, application of optimisation as well as design and sensitivity analyses, linking requirement and customer-based design and simulation as well as the data management of modelling and simulation. The project was also funded by several companies.
• Tekes-funded project HUBI, 2011-2013, (joint project with VTT) developed tools for multilevel modelling and simulation of process plants and their engineering and business processes. The tools were to be connected to a common use environment (hub) through which the models can be made available also for other users. In addition to the solvers of different levels of details, tools for optimization were studied. The project was also funded by several companies.
• Academy of Finland funded the project Strategic Development of Multiobjective Optimization: Theory and Software in 2009-2012. The project focused on both theoretical challenges of multiobjective optimization and software implementation of IND-NIMBUS.
• ForestCluster (a company owned by several universities, research institutes and firms in the forest and pulp and paper sector) and Tekes funded a project POJo and its continuation project WP9 in the EffNet program of Finnish Bioeconomy Cluster Ltd (former Forestcluster Ltd). POJo was a joint project with Tampere University of Technology, VTT, Helsinki University of Technology and University of Kuopio in 2008-2010. WP9 was a joint project with Tampere University of Technology, VTT, Aalto University and University of Eastern Finland in 2010-2012. In these projects, a new model-based and optimizing design concept for material and information flows in production systems was developed. The overall aim was to increase flexibility in process design and reduce the amount of capital invested in production lines. The main application area was pulp and paper industry.
• Tekes funded the project BioScen (joint project with Aalto University and the Technical Research Centre of Finland (VTT)), which was a part of the BioRefine technology programme in 2008-2011. The research was focused on developing a basis for the modeling and simulation of the unit processes of a biorefinery. The sensitivity of such models to uncertainties in process parameters, optimization of the production plant concepts and finally for life-cycle analysis of the biorefinery products were considered. This project was also funded by several companies.
• University Alliance funded the project Measurements, Data Analysis and Multiobjective Optimization (MeMO) in 2008-2010.
• Tekes funded the project Hyvä-tietää, multiobjective optimization and multidisciplinary decision support (joint project with Helsinki School of Economics, University of Kuopio, Tampere University of Technology and Helsinki University of Technology), which belonged to the MASI technology programme, in 2005-2008. Several companies were involved.
• Tekes funded the project NIMBUS - multiobjective optimization in product development in 2002-2005. Several companies funded the research as well. Further information is available at the project website.
• Several projects of the Academy of Finland devoted to nonlinear multiobjective optimization and multiple criteria decision making (method, theory and software development) have been active during the years.

For further information, see publications of the Multiobjective Optimization Group.