Bhupinder Singh Saini
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Bhupinder Singh Saini has a total of 31 co-authors. Here are the top co-authors by number of shared publications:
- Kaisa Miettinen
University of Jyväskylä, Faculty of Information Technology, Finland
11 shared publications - Babooshka Shavazipour
University of Jyväskylä
7 shared publications - Francisco Ruiz
Department of Applied Economics (Mathematics), Universidad de Málaga, Málaga, Spain
2 shared publications - Giomara Lárraga
University of Jyväskylä, Faculty of Information Technology, Finland
2 shared publications - Juuso Pajasmaa
University of Jyvaskyla, Faculty of Information Technology, University of Jyvaskyla, Finland
2 shared publications Kaisa Miettinen
University of Jyväskylä, University of Jyväskylä, Finland
2 shared publications
Latest Publications
- 23 August 2026
- Published by Springer Nature
- p. 350-366
Preference Guided Multiobjective Bayesian Optimization with Aspiration and Reservation Levels
- 23 August 2026
- Published by Springer Nature
- p. 350-366
Preference Guided Multiobjective Bayesian Optimization with Aspiration and Reservation Levels
- Open Access
- 27 May 2026
- Published by Springer Nature in Journal of Global Optimization
- Vol. 95 (2) , 383-404
NAUTILI: A trade-off-free interactive multiobjective optimization method for group decision making
We consider problems with multiple conflicting objective functions and multiple decision makers and concentrate on interactive methods, where the preferences of the decision makers are iteratively incorporated in the solution process. We propose a novel trade-off-free interactive multiobjective optimization method NAUTILI for group decision making. In NAUTILI, the solution process starts from an inferior solution. Then, each decision maker individually interacts with NAUTILI via specialized visualizations in a graphical user interface to provide their preferences. NAUTILI determines the direction of simultaneous improvements for the whole group. Thus, in each iteration, all decision makers gain in all objective functions until reaching the most preferred solution by the group. This solution is so-called Pareto optimal. The method can be seen as a win-win approach, as in each iteration, all the objective functions can be improved simultaneously for all decision makers. As a proof of concept, we demonstrate the potential and benefits of NAUTILI with a forest management problem involving three decision makers. We also asked the decision makers to answer a questionnaire about their satisfaction with the final solution, the method’s user-friendliness, and how they felt that their preferences were reflected in the solution process. The findings support the applicability of the method. This supports the potential of applying an interactive and trade-off-free method to group decision making in multiobjective optimization.
- 27 May 2026
- Published by Springer Nature in Journal of Global Optimization
- Vol. 95 (2) , 383-404
Open AccessNAUTILI: A trade-off-free interactive multiobjective optimization method for group decision making
We consider problems with multiple conflicting objective functions and multiple decision makers and concentrate on interactive methods, where the preferences of the decision makers are iteratively incorporated in the solution process. We propose a novel trade-off-free interactive multiobjective optimization method NAUTILI for group decision making. In NAUTILI, the solution process starts from an inferior solution. Then, each decision maker individually interacts with NAUTILI via specialized visualizations in a graphical user interface to provide their preferences. NAUTILI determines the direction of simultaneous improvements for the whole group. Thus, in each iteration, all decision makers gain in all objective functions until reaching the most preferred solution by the group. This solution is so-called Pareto optimal. The method can be seen as a win-win approach, as in each iteration, all the objective functions can be improved simultaneously for all decision makers. As a proof of concept, we demonstrate the potential and benefits of NAUTILI with a forest management problem involving three decision makers. We also asked the decision makers to answer a questionnaire about their satisfaction with the final solution, the method’s user-friendliness, and how they felt that their preferences were reflected in the solution process. The findings support the applicability of the method. This supports the potential of applying an interactive and trade-off-free method to group decision making in multiobjective optimization.
- Open Access
- 2 May 2026
- Published by Springer Nature in Journal of Global Optimization
- Vol. 95 (4) , 1-20
Benefits of the IOPIS paradigm on the landscapes of multiobjective optimization problems
The preference incorporated space (IOPIS) paradigm in interactive multiobjective optimization introduces a transformation or mapping of the objective space to a space where the coordinates are the values of scalarization functions. The scalarization functions include preference information provided by a decision maker. Therefore, IOPIS provides new possibilities for solving multiobjective optimization problems. In this paper, we investigate the effects of the transformation on the landscapes of multiobjective optimization problems by altering the landscape and the distribution of locally Pareto optimal solutions. The analysis includes both theoretical proofs and illustrations showcasing the landscapes of multiobjective optimization problems. Theoretical results establish that the transformation using achievement scalarizing functions does not introduce any new locally Pareto optimal solutions, i.e., if a solution is not locally Pareto optimal for the original problem, it cannot become locally Pareto optimal after the transformation. Moreover, we derive conditions under which local Pareto optima in the original problem are preserved and derive a geometrical description of the region where local optima can be found. Visualizations illustrate the theoretical findings for a number of examples. Besides its advantages in interactive multiobjective optimization, the IOPIS transformation may also help “smooth out” local optima in conventional a posteriori methods while retaining global optima.
- 2 May 2026
- Published by Springer Nature in Journal of Global Optimization
- Vol. 95 (4) , 1-20
Open AccessBenefits of the IOPIS paradigm on the landscapes of multiobjective optimization problems
The preference incorporated space (IOPIS) paradigm in interactive multiobjective optimization introduces a transformation or mapping of the objective space to a space where the coordinates are the values of scalarization functions. The scalarization functions include preference information provided by a decision maker. Therefore, IOPIS provides new possibilities for solving multiobjective optimization problems. In this paper, we investigate the effects of the transformation on the landscapes of multiobjective optimization problems by altering the landscape and the distribution of locally Pareto optimal solutions. The analysis includes both theoretical proofs and illustrations showcasing the landscapes of multiobjective optimization problems. Theoretical results establish that the transformation using achievement scalarizing functions does not introduce any new locally Pareto optimal solutions, i.e., if a solution is not locally Pareto optimal for the original problem, it cannot become locally Pareto optimal after the transformation. Moreover, we derive conditions under which local Pareto optima in the original problem are preserved and derive a geometrical description of the region where local optima can be found. Visualizations illustrate the theoretical findings for a number of examples. Besides its advantages in interactive multiobjective optimization, the IOPIS transformation may also help “smooth out” local optima in conventional a posteriori methods while retaining global optima.
- Open Access
- 8 April 2026
- Published by Elsevier in Forest Policy and Economics
- Vol. 187
Advancing climate-smart forestry with an interactive decision-support tool that incorporates the preferences of private forest owners
- 8 April 2026
- Published by Elsevier in Forest Policy and Economics
- Vol. 187
Open AccessAdvancing climate-smart forestry with an interactive decision-support tool that incorporates the preferences of private forest owners
- Open Access
- 30 January 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Evolutionary Learning and Optimization
- Vol. 6 (3) , 1-51
D-PHI: Desirability-Based Hypervolume Indicator for Interactive Multiobjective Optimization Using Aspiration and Reservation Levels as Preferences
To address problems with multiple conflicting objective functions while incorporating preference information from a decision maker (a domain expert), interactive evolutionary multiobjective optimization methods have been proposed and widely adopted. To systematically assess the performance of these methods, it is desirable to have indicators that not only rely on the Pareto dominance relation, where solutions with better objective function values are preferred, but also take the decision maker's preferences into account. In this way, indicator values can meaningfully reflect the decision maker's preferences and the quality of the solutions obtained. In this paper, we propose the first performance indicator that incorporates preference information in the form of two reference points, specifically, an aspiration point, consisting of aspiration levels that represent the decision maker's full satisfaction, and a reservation point, consisting of reservation levels that represent dissatisfaction. The new indicator, termed D-PHI, translates information about reservation and aspiration points into a transformed hypervolume measure that assesses the performance of a solution set. We introduce D-PHI with an extended set of guidelines for preference-based performance evaluation, discuss its theoretical properties, and demonstrate its behavior through empirical examples. Moreover, we define a complementary indicator (CI) based on a so-called achievement scalarizing function. CI offers additional insight into whether aspiration and reservation levels have been achieved, and its usefulness is demonstrated through examples.
- 30 January 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Evolutionary Learning and Optimization
- Vol. 6 (3) , 1-51
Open AccessD-PHI: Desirability-Based Hypervolume Indicator for Interactive Multiobjective Optimization Using Aspiration and Reservation Levels as Preferences
To address problems with multiple conflicting objective functions while incorporating preference information from a decision maker (a domain expert), interactive evolutionary multiobjective optimization methods have been proposed and widely adopted. To systematically assess the performance of these methods, it is desirable to have indicators that not only rely on the Pareto dominance relation, where solutions with better objective function values are preferred, but also take the decision maker's preferences into account. In this way, indicator values can meaningfully reflect the decision maker's preferences and the quality of the solutions obtained. In this paper, we propose the first performance indicator that incorporates preference information in the form of two reference points, specifically, an aspiration point, consisting of aspiration levels that represent the decision maker's full satisfaction, and a reservation point, consisting of reservation levels that represent dissatisfaction. The new indicator, termed D-PHI, translates information about reservation and aspiration points into a transformed hypervolume measure that assesses the performance of a solution set. We introduce D-PHI with an extended set of guidelines for preference-based performance evaluation, discuss its theoretical properties, and demonstrate its behavior through empirical examples. Moreover, we define a complementary indicator (CI) based on a so-called achievement scalarizing function. CI offers additional insight into whether aspiration and reservation levels have been achieved, and its usefulness is demonstrated through examples.
- 1 January 2026
- Published by Canadian Science Publishing in Canadian Journal of Forest Research
- Vol. 56, 1-13
Searching for sustainable national forest harvest levels through interactive multiobjective optimization
In Finland, linear optimization has been used to search for sustainable national harvest levels for over 30 years—commonly maximizing net present value from forests county by county, while ensuring sustainability of wood production by requiring nondecreasing harvest levels over time. However, this approach does not help policymakers understand how the selected harvest level is reached and what affects the result. Moreover, optimizing per county may lead to an inefficient solution, as the counties serve as constraints. National-level interactive planning is a better approach, but such interactive optimization is compute-intensive. We present an interactive multiobjective optimization approach applying the NAUTILUS Navigator method. It is based on precalculated optimal solutions and gradual improvements from the nadir (the worst solution). We apply the approach in a novel way: starting with a sparse representation of optimal solutions and generating a denser representation in the area of interest, saving computational resources. The problem formulation is also improved during the solution process. Moreover, the interactive approach advises decision makers on the consequences of the selected solution better than a few preselected scenarios. The decision maker can direct the solution process and find preferred solutions. The results also show improvement in solutions when the county-level constraints are removed.
- 1 January 2026
- Published by Canadian Science Publishing in Canadian Journal of Forest Research
- Vol. 56, 1-13
Searching for sustainable national forest harvest levels through interactive multiobjective optimization
In Finland, linear optimization has been used to search for sustainable national harvest levels for over 30 years—commonly maximizing net present value from forests county by county, while ensuring sustainability of wood production by requiring nondecreasing harvest levels over time. However, this approach does not help policymakers understand how the selected harvest level is reached and what affects the result. Moreover, optimizing per county may lead to an inefficient solution, as the counties serve as constraints. National-level interactive planning is a better approach, but such interactive optimization is compute-intensive. We present an interactive multiobjective optimization approach applying the NAUTILUS Navigator method. It is based on precalculated optimal solutions and gradual improvements from the nadir (the worst solution). We apply the approach in a novel way: starting with a sparse representation of optimal solutions and generating a denser representation in the area of interest, saving computational resources. The problem formulation is also improved during the solution process. Moreover, the interactive approach advises decision makers on the consequences of the selected solution better than a few preselected scenarios. The decision maker can direct the solution process and find preferred solutions. The results also show improvement in solutions when the county-level constraints are removed.
- 28 February 2025
- Published by Springer Nature
- p. 241-256
An Efficient Iterative Approach for Uniformly Representing Pareto Fronts
- 28 February 2025
- Published by Springer Nature
- p. 241-256
An Efficient Iterative Approach for Uniformly Representing Pareto Fronts
- Open Access
- 4 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 164371-164388
SCORE Band Visualizations: Supporting Decision Makers in Comparing High-Dimensional Outcome Vectors in Multiobjective Optimization
Clearly arranged visualizations are needed in multiobjective optimization problems with a large number of objective functions, when a large number of Pareto optimal outcome vectors (vectors of objective function values) must be compared during the decision making processes. This paper contributes to visualizing such outcome vectors independent of how they have been generated. Parallel coordinate plots are a widely used visualization technique to represent different outcome vectors. We propose a novel visualization technique called SCORE bands to be used with parallel coordinate plots to support the decision maker in simultaneously identifying patterns in outcome vectors and correlations among the objective functions in a meaningful way. To do so, amongst others, we change the ordering of objective functions and modify the distances among them in parallel coordinate plots. SCORE bands also have interactive capabilities allowing the decision maker to first study general trends among the outcome vectors as bands and then zoom-in and move about different groups of outcome vectors of interest. The novelty of our approach lies in proposing a visually appealing way to support the decision maker in dealing with large amounts of information. We demonstrate the benefits of SCORE bands with different examples.
- 4 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 164371-164388
Open AccessSCORE Band Visualizations: Supporting Decision Makers in Comparing High-Dimensional Outcome Vectors in Multiobjective Optimization
Clearly arranged visualizations are needed in multiobjective optimization problems with a large number of objective functions, when a large number of Pareto optimal outcome vectors (vectors of objective function values) must be compared during the decision making processes. This paper contributes to visualizing such outcome vectors independent of how they have been generated. Parallel coordinate plots are a widely used visualization technique to represent different outcome vectors. We propose a novel visualization technique called SCORE bands to be used with parallel coordinate plots to support the decision maker in simultaneously identifying patterns in outcome vectors and correlations among the objective functions in a meaningful way. To do so, amongst others, we change the ordering of objective functions and modify the distances among them in parallel coordinate plots. SCORE bands also have interactive capabilities allowing the decision maker to first study general trends among the outcome vectors as bands and then zoom-in and move about different groups of outcome vectors of interest. The novelty of our approach lies in proposing a visually appealing way to support the decision maker in dealing with large amounts of information. We demonstrate the benefits of SCORE bands with different examples.
- Open Access
- 15 July 2023
- Published by Association for Computing Machinery (ACM)
- p. 1703-1711
Using a Database to Support Interactive Multiobjective Optimization, Visualization, and Analysis
Many libraries of open-source implementations of multiobjective optimization problems (MOPs) and evolutionary algorithms (MOEAs) have been developed in recent years. These libraries enable researchers to solve their MOPs using diverse MOEAs. Some libraries also implement interactive MOEAs, which enable decision-makers (experts in the domain of the MOP) to provide their preferences and guide the optimization process toward their region of interest. These libraries also provide access to visualization methods and benchmarking tools. However, they do not currently implement a database to store and utilize the data generated while running MOEAs. We propose the creation of SIVA DB, a database designed to be easily incorporated into existing libraries as a modular addition. SIVA DB provides a standard way to archive an MOEA's population and the metadata associated with each population member. Such metadata can include, e.g., the parameters and state of the MOEA and the preferences the decision-maker gives (in the case of interactive MOEAs). The database can store data from multiple runs of any number of MOEAs, and even data from different MOPs. SIVA DB provides easy access to the contained data to analyze the optimization process or create efficient MOEAs. We demonstrate the latter in this paper with experiments.
- 15 July 2023
- Published by Association for Computing Machinery (ACM)
- p. 1703-1711
Open AccessUsing a Database to Support Interactive Multiobjective Optimization, Visualization, and Analysis
Many libraries of open-source implementations of multiobjective optimization problems (MOPs) and evolutionary algorithms (MOEAs) have been developed in recent years. These libraries enable researchers to solve their MOPs using diverse MOEAs. Some libraries also implement interactive MOEAs, which enable decision-makers (experts in the domain of the MOP) to provide their preferences and guide the optimization process toward their region of interest. These libraries also provide access to visualization methods and benchmarking tools. However, they do not currently implement a database to store and utilize the data generated while running MOEAs. We propose the creation of SIVA DB, a database designed to be easily incorporated into existing libraries as a modular addition. SIVA DB provides a standard way to archive an MOEA's population and the metadata associated with each population member. Such metadata can include, e.g., the parameters and state of the MOEA and the preferences the decision-maker gives (in the case of interactive MOEAs). The database can store data from multiple runs of any number of MOEAs, and even data from different MOPs. SIVA DB provides easy access to the contained data to analyze the optimization process or create efficient MOEAs. We demonstrate the latter in this paper with experiments.
- 9 March 2023
- Published by Springer Nature
- p. 578-592
Incorporating Preference Information Interactively in NSGA-III by the Adaptation of Reference Vectors
- 9 March 2023
- Published by Springer Nature
- p. 578-592
Incorporating Preference Information Interactively in NSGA-III by the Adaptation of Reference Vectors