WP 3 Decision-making, potential, and business models

lead by Dr. Mikael Collan, Lappeenranta University of Technology (LUT)
[email protected]
+358 505567185

The newest MFG4.0 dissertation introduces nearest neighbor-based novel soft computing techniques for supervised classification and regression applications

The newest MFG4.0 dissertation introduces nearest neighbor-based novel soft computing techniques for classification and regression applications

This dissertation focuses on supervised machine learning techniques – classification and regression. In particular, the emphasis is on the fuzzy k-nearest neighbor (FKNN) algorithm that has received substantial attention in classification problems due to its efficacy and flexibility.

In classification, learning from data can be challenging for many algorithms due to uncertainties and inconsistencies in the data. A typical issue associated with most classification problems is that class distributions in the data are imbalanced – meaning data points do not equally represent the classes in a class variable, which can significantly affect classification performance. Given this issue, this research develops solution techniques based on the FKNN algorithm.

The first proposed approach is the multi-local Power mean fuzzy k-nearest neighbor (MLPM-FKNN), which uses class prototype local mean vectors instead of individual training instances in the learning part. The second proposed method is the Bonferroni mean-based fuzzy k-nearest neighbor (BM-FKNN), an enhancement of the MLPM-FKNN method using the Bonferroni mean to compute class prototype local mean vectors. The findings of the tests with different real-world data sets demonstrate the robustness and efficacy of the proposed approaches for class imbalance problems.

Figure: Kumbure, M. M., Luukka, P., & Collan, M. (2020) A new fuzzy k-nearest neighbor classifier based on the Bonferroni mean. Pattern Recognition Letters, 140, 172-178

Mahinda Mailagaha Kumbure

Junior Researcher, LUT University

In addition to the classification context, the rationale underlying the FKNN algorithm for regression problems is successfully generalized, and concurrently, a novel Minkowski distance-based fuzzy k-nearest neighbor (Md-FKNNreg) method is proposed. This method is tested on several different real-world data sets, and the results show that it achieves the best performance compared to the several state-of-the-art regression methods.

Mahinda Mailagaha Kumbure defended his dissertation in the field of Business and Management (Business Analytics) at LUT University on the 10th of November 2022. The title of his doctoral thesis is Novel fuzzy k-nearest neighbor methods for effective classification and regression.

Read more

Mailagaha Kumbure, Mahinda (2022-11-10) Novel fuzzy k-nearest neighbor methods for effective classification and regression


machine learning, classification, regression, feature selection, prediction, class imbalance, fuzzy k-nearest neighbor, local means, performance

Towards sustainable society -6 different viewpoints of  the MFG4.0 research presented at Strategic Research Scientific Conference

Towards sustainable society -6 different viewpoints of  the MFG4.0 research presented at Strategic Research Scientific Conference

The Strategic Research conference brings together scholars from different research fields with a common interest in exploring responses to societal challenges in a fair, just, and sustainable society. The conference will foster an exchange of ideas, approaches, and insights between the disciplines.


Strategic Research – Scientific Conference:
 A fair, just and sustainable society


12 October 2022 – 13 October 2022

Epicenter, Mikonkatu 9, Helsinki

Our researchers will present abstracts on the following themes:

Additive Manufacturing Point-of-View to Firm Resilience

Presenter:Mikael Collan

Resilience is the ability of firms to cope with sudden and dramatic changes in the business environment. Typically, firms with the flexibility to change how they operate in terms of what they produce and how they work are less vulnerable to dramatic changes than companies with fixed architectures and product assortments.

This presentation talks about the manufacturing and national resilience that is created through added manufacturing. When fleets of additive manufacturing equipment are harnessed to provide critical manufacturing in times of crisis, they are a source of resilience. This requires preparation and active orchestration.

Mikael Collan

Professor at LUT University, Director General at VATT (Institute for Economic Research)

Towards sustainability in the metal industry using 3D printing

Presenter: Jyrki Savolainen
Co-authors: Ilkka Poutiainen, Marika Hirvimäki, Kari Ullakko, Ville Laitinen

The interest in additive manufacturing (AM), commonly known as 3D printing, has grown enormously over the last decade. AM is a novel way to produce unique parts offering new business opportunities and improved environmental sustainability of production. The current literature on AM emphasizes its positive effects on the sustainability of global production supply chains. We claim that the AM design and production can either positively or negatively influence the final product’s overall cost and resource efficiency.

Metal powders for AM are expensive, and manufacturing these powders consumes significantly more energy than sheet or bar materials. In this paper, we discuss how cost-effectiveness and sustainability are achieved when parts are redesigned using, e.g., lattice structures, an optimal printing position, and minimizing the need for post-processing.

Jyrki Savolainen

Post Doctoral Researcher at LUT University

Towards General Theory of Sustainable Development: Systematic Digital Twin Operationalisation of the Grand Sustainability Theory

Presenter: Jari Kaivo-oja

Co-authors: Jyrki Luukkanen, Jarmo Vehmas & Tadht O ́MahonyAcross decades of contemporary discussion on sustainable development, a core debate has concerned whether economic growth can be made sustainable, environmentally and socially.
In recent years, this has become a debate about economic growth versus economic degrowth, whether the former can be environmentally sustainable and whether the latter can be socially sustainable. In the presentation, the author presents the general theory of sustainable growth and development and links the use of the model to the development challenges of digital twins.

This developed theory can be used in national sustainability assessment and planning, as well as in regional policy or urban planning. The model forms the core of sustainable development or sustainability science.

Jari Kaivo-oja

Research Director, Dr, Adjunct Professorat University of Turku

Towards a Sustainable Portfolio Theory - Foresight-driven strategic asset allocation and SDG (Sustainable Development Goals) alignment

Presenter: Ville Korpela
Co-authors: Jari Kaivo-oja, Arne Fagerström & Petri Kuusisto

Modern Portfolio Theory (MPT) has formed the core theoretical basis for asset allocation strategies of institutional investors for several decades. Recently, there have been wider calls among practitioners and the academic community to extend the existing theory to include broader impacts on the societal level to align investment practices globally with the Sustainable Development Goals (SDGs). In the paper, we discuss an outline for a sustainable portfolio theory (SPT) by extending the existing modern portfolio theory (MPT). Our objective is to demonstrate a relationship between the values businesses create and the impact investors are capturing.

Ville Korpela

PhD Researcher, Turku School of Economics, University of Turku

Visionary-Thinking in Sustainable Industrial Development in Finland - The Case of Allied ICT Finland ́s Vision-Building Processy

Presenter: Mikkel Knudsen
Co-authors: Jari Kaivo-oja & Tero Villman

Europe must renew itself to retain its 35 million industrial jobs in an increasingly fierce global competition. The industrial renewal must happen in concerto with bold ambitions for the twin, green and digital transitions, and it must be executed within the modern VUCA (volatility, uncertainty, complexity, and ambiguity) decision environment. Our research highlights potential pathways for Finnish industry and steps for the Finnish government, businesses, and society. Based on the visioning process, Industry 6.0 is defined as “ubiquitous, customer-driven, virtualized, antifragile manufacturing”. It is characterized by customer-centric, highly customized lot-size-1 thinking and by hyper-connected factories with dynamic supply chains and data flows across domains.

Mikkel Knudsen

Project Researcher, Finland Futures Research Centre (FFRC) M.Sc. (Pol. Science) University of Turku

The challenges of long-term development for Finnish education and education policy to meet the demands of Manufacturing 4.0

Presenter: Maarit Virolainen
Co-author: Juhani Rautopuro

The so-called Manufacturing 4.0, sets needs to reform education, and they relate to the following three points: (1) to define, picture, and design what kind of reforms and for which parts of the education system and curricula are needed to instil fair, just and sustainable society and adapt to the change of technology in society, (2) to assess and evaluate, if the targeted change has been met in the learning of various learner groups before and after the reform. The presentation discusses the latest reforms in education and the role of research in identifying developmental education needs.

Maarit Virolainen

Project Researcher PhD (Adult education), MA (Social sciences) at University of Jyväskylä

The ‘Pyramid model’ helps to tune your organization and business models to take advantage of technological innovations

The ‘Pyramid model’ helps to tune your organization and business models to take advantage of technological innovations

Monetization of technological innovation necessitates adjustments to the organization, business model(s), and even ecosystems. How to make money with the innovation or how it contributes to profit generation must be clear. If innovation activity has been business need-driven, you might already have an idea of that. If innovation has been technology-driven i.e., you have developed or tested new technology to understand what could be done, you need to form an understanding of how to monetize it in your case.

Mikko Hirvonen

Strategy advisor & Researcher,
LUT University

Regardless of the driver of innovation, the prerequisite for successful monetization is an alignment of a business strategy and business model(s). When strategy and business model choices are done, it’s time to look at the ‘set of organizational capabilities’ needed to run the business and to meet the strategic objectives. For example, in a case of product innovation, you might need new or adjusted capabilities to commercialize and manufacture it, and to ensure that the customer gains the expected value. At the same time, you and your business partners need to be able to capture enough value that motivates to continue doing business.

The ‘Pyramid model’ is a framework to support this – often quite complex task to realize business benefits from different types of innovation (be it a product, process, organization, or business model innovation) and to monetary value of it. With the ‘Pyramid model’ the organization and its business model(s), where competitive advantage is potentially created, are broken down to a set of organizational capabilities and further into interconnected capability elements.

Depending on the type of innovation, it changes one or more of the capability elements. The ‘Pyramid model helps to identify required adjustments to interdependent capability elements to achieve targeted capability with expected outcomes, but also interconnected capabilities to gain business benefits on business model level and to ultimately monetize the innovation.

The ‘Pyramid model’ is expected to be of significant value to the practitioners when directing innovation efforts and designing value configuration, and organizations and ecosystems around it. For researchers it provides a novel framework for research serving the practice.

Hirvonen, Mikko H. (2022). Pyramid model – Conceptualizing an organizational capability to design IT investments. Proceedings of the 15th IADIS International Conference Information Systems. Virtual Conference 12-14 March 2022.  


Pyramid model, business models, business model innovation, business model framework, capabilities, capability system

How to make decisions on complex environmental issues?

How to make decisions on complex environmental issues?

In practice, environmental analytics involves an integration of science, methods, and techniques involving a combination of computers, computational intelligence, information technology, mathematical modelling, and system science to address “real-world” environmental and sustainability problems. Effective environmental decision-making is often challenging and complex where final results often involve inherently subjective political and socio-economic facets. Furthermore, while certain environmental and sustainability decision-making specifications may be self-evident (post hoc analysis always tends to be incredibly accurate), more typical problems possess components that cannot be directly included in the underlying decision process without additional manipulation. Such decision-making is frequently further compounded by additional stochastic uncertainties. Consequently, complex “real world” sustainability problems frequently employ computational decision-making approaches to construct solutions to applications containing numerous quantitative dimensions and considerable sources of uncertainty.

Mariia Kozlova

Post-doctoral Researcher, Docent,
LUT University

Julian Scott Yeomans

Professor, Operations Management and Information Systems,
Director, Master of Business Analytics (MBAN) Program,
Director, Master of Management in Artificial Intelligence (MMAI) Program,
Schulich School of Business,
York University, Canada

We trust that the number and quality of the papers will prove to be of significant value to the many different researchers and practitioners who actively engage in applying disparate computational methodologies to sustainability analysis and environmental decision-making using simulation, optimization, and analytics. It is our sincere hope that this book will not only enlighten readers on the current state-of-the-art applications in computational sustainability, but will also serve to inspire further collaboration and cooperation on extensions to these topics. Continuing advancement on these topics is always necessary as “It is difficult to predict, especially the future” (Danish proverb often attributed to Niels Bohr) but, more to the point and borrowing from the deeply philosophical characters in the cartoon Calvin and Hobbes, “The trouble with the future is that it keeps turning into the present”. 

The book by Mariia Kozlova and Julian Scott YeomansSustainability Analysis and Environmental Decision-Making Using Simulation, Optimization, and Computational Analytic” includes a number of such applied computational analytics contributions that either create new decision-making methods or provide innovative implementations of existing methods for addressing a wide spectrum of sustainability applications, broadly defined. The rich diversity of applications within the papers exemplifies the considerable range of both methodological relevance and practical contributions to research in environmental analysis. The disparate contributions all emphasize novel approaches of computational analytics as applied to environmental decision-making and sustainability analysis –be this on the side of optimization, simulation, modelling, computational solution procedures, visual analytics, and/or information technologies.


from Mariia Kozlova, Timo Nykänen and Julian Scott Yeomans Technical Advances in Aviation Electrification: Enhancing Strategic R&D Investment Analysis through Simulation Decomposition Reprinted from: Sustainability 2022, 14, 414, doi:10.3390/su14010414


environmental decision-making; simulation; simulation decomposition; optimization; computational analytics; visual analytics; sustainability analysis; waste management; water resource planning; energy policy; climate change; industrial ecology; resource recovery; recycling

Pandemiaan liittyvillä sosiaalipoliittisilla toimilla myös ennaltaehkäistiin uusien riskien toteutumista

Pandemiaan liittyvillä sosiaalipoliittisilla toimilla myös ennaltaehkäistiin uusien riskien toteutumista

Juuri julkaistussa raportissa selvitettiin  pandemiaan liittyviä sosiaalipoliittisia toimia Suomessa ja 12 muussa OECD-maassa vuonna 2020.

Lisäksi analysoitiin, millaisia muutoksia toimet olivat suhteessa pandemiaa edeltäviin politiikkatoimiin. Tutkimusaineisto koottiin ensisijaisesti hallitusten, parlamenttien ja ministeriöiden dokumenteista.

Sosiaalipoliittisissa muutoksissa korostuivat etuuksien tasokorotukset, omavastuupäivien poistot ja etuusjaksojen pidennykset. Suomessa, Ruotsissa ja Norjassa tämänkaltaisten muutosten suhteellinen osuus oli suurin. Sen sijaan viimesijaista sosiaaliturvaa muutettiin vähiten. Sosiaaliturvaa laajennettiin 11 maassa. Uusista toimista yleisimpiä olivat kerta­luontoiset tulonsiirrot kahdeksassa maassa.

Päivi Mäntyneva

Helsingin yliopisto

Johanna Peltoniemi

Helsingin yliopisto

Eeva-Leena Ketonen

Helsingin yliopisto

Henri Aaltonen

tekninen avustaja,
Helsingin yliopisto

Heikki Hiilamo

Helsingin yliopisto

Julkisessa keskustelussa eri maiden sosiaalipoliittiset toimet esitettiin erityisesti kriisi- ja hätäapuna keskellä pandemiaa. Vastoin yleistä mielikuvaa eri maiden sosiaalipoliittiset toimet eivät olleet pääasiassa korjaavia toimia ja reagointia jo toteutuneisiin riskeihin. Päinvastoin, sosiaalipoliittiset toimet olivat ennemmin uusia riskejä ennaltaehkäiseviä toimia. Koko tutkimusaineistossa ennaltaehkäisevien toimien osuus oli 55 prosenttia (N=113).

Tutkimustulosten mukaan enemmistö (76 %) tehdyistä toimista oli muutoksia jo olemassa oleviin etuusjärjestelmiin, kuten etuuksien korotuksiin ja etuusjaksojen pidennyksiin. Kaikki muutokset olivat väestön kannalta parannuksia olemassa oleviin etuuksiin; emme havainneet ainuttakaan tapausta etuuksien leikkauksista. Parannukset olemassa oleviin etuuksiin näkyivät etenkin Pohjois-maissa ja Suomessa, Ruotsissa ja Norjassa, joissa yli 90 prosenttia toimista lukeutui tähän ryhmään.

Suomessa toteutetut sosiaalipoliittiset toimet liittyivät työttömyysturvaan, sairausvakuutukseen, lapsiperheiden etuuksiin, viimesijaisiin etuuksiin, työllisyyden edistämiseen ja ylivelkaantumisen ehkäisyyn. Enemmistö toimista oli ennaltaehkäiseviä. Hyvinvointivaltioregiimeittäin tarkasteltuna ennaltaehkäisevien toimien osuus oli suurempi verrattuna korjaaviin toimiin – vaikkakin erot olivat melko pienet. 

Kriisin aikaisiin sosiaalipoliittisiin toimiin ja niiden toimeenpanoon vaikuttivat lukuisat tekijät. Kansallinen kehitys oli yhteydessä myös kansainvälisiin vaikutteisiin. Tällä  puolestaan on merkitystä sosiaalipoliittisten järjestelmien kehittämiseen. Sosiaalipoliittisten ideoiden nähtiin leviävän maasta toiseen. Pandemian aikana vertaisoppiminen – tai ainakin toimien samankaltaisuudet – oli nähtävissä tämän tutkimuksen tuloksissa etenkin työllisyyden edistämistoimina, lomautusjärjestelmien ja palkkatuen laajentamisena sekä erilaisina työajan lyhennysohjelmina, joiden avulla voidaan lievittää ja ennaltaehkäistä sosiaalisia ja taloudellisia ongelmia ja eriarvoistumista.

Tutkimuksessa oli mukana yhteensä 208 sosiaalipoliittista tointa regiimiteorian pohjalta valituissa maissa: Suomi, Norja, Ruotsi, Tanska, Islanti, Alankomaat, Saksa, Espanja, Italia, Iso-Britannia, Yhdysvallat, Etelä-Korea ja Japani. Toimet liittyivät työttömyysturvaan, sairausvakuutukseen, lapsiperheiden etuuksiin, eläkkeisiin, viimesijaisiin tukiin, suoriin tulonsiirtoihin, työllisyyden edistämiseen, asumisen tukemiseen, opiskelijoiden etuuksiin ja ylivelkaantuneisuuden ehkäisyyn.

Mäntyneva, P., Ketonen, E.-L., Peltoniemi, J., Aaltonen, H. & Hiilamo, H. (2021) Sosiaalipoliittiset toimet koronapandemian aikana vuonna 2020 : Vertailututkimus Suomesta ja 12 muusta OECD-maasta. Sosiaaliturvakomitean julkaisuja 2021:2

Summary in English:
The study examines what social policy measures related to the COVID-19 pandemic Finland and 12 other OECD countries implemented in 2020. It also analyses the changes in the social policy measures in relation to pre-pandemic policy measures. The data for this study was collected primarily from documents of governments, parliaments and ministries. The study analyses a total of 208 social policy measures in the following countries selected on the basis of regime theory: Finland, Norway, Sweden, Denmark, Iceland, the Netherlands, Germany, Spain, Italy, the United Kingdom, the United States, South Korea and Japan.

The findings of this study show that the measures were related to unemployment benefits, health insurance cash benefits, benefits for families with children, pensions, last-resort assistance, direct income transfers, employment promotion, housing support, student benefits and prevention of over-indebtedness.

The social policy changes focused on increases in benefits, the removal of the waiting period for unemployment benefits and the extension of benefit periods. The relative share of such changes was the highest in Finland, Sweden and Norway, whereas the last-resort form of social security was changed the least in all countries studied. Social security coverage was extended in 11 countries. One-off income transfers were the most common new measures in eight countries. The social policy measures used due to the COVID-19 pandemic were not only reactive but they also prevented the emergence of new risks.


koronakriisi, sosiaalipolitiikka, sosiaaliturvan muutokset, OECD-maat, sosiaaliturvauudistus


COVID-19 crisis, social policy, changes in social security, OECD countries, social security reform