Home Menu
icon-menu-close.svg

Faculty & Research

Faculty & Research

Research Productive

Show result

Clear
Search Query :
Area :
Search Query :
626 items in total found

Journal Articles | 2017

A review on bilevel optimization: from classical to evolutionary approaches and applications

Ankur Sinha, Pekka Malo, and Kalyanmoy Deb

IEEE Computational Intelligence Society

Bilevel optimization is defined as a mathematical program, where an optimization problem contains another optimization problem as a constraint. These problems have received significant attention from the mathematical programming community. Only limited work exists on bilevel problems using evolutionary computation techniques; however, recently there has been an increasing interest due to the proliferation of practical applications and the potential of evolutionary algorithms in tackling these problems. This paper provides a comprehensive review on bilevel optimization from the basic principles to solution strategies; both classical and evolutionary. A number of potential application problems are also discussed. To offer the readers insights on the prominent developments in the field of bilevel optimization, we have performed an automated text-analysis of an extended list of papers published on bilevel optimization to date. This paper should motivate evolutionary computation researchers to pay more attention to this practical yet challenging area.

Go to link
Read More

Journal Articles | 2017

Optimal management of naturally regenerating uneven-aged forests

Ankur Sinha, Janne Ramo, Pekka Malo, and Olli Tahvonen

European Journal of Operational Research

A shift from even-aged forest management to uneven-aged management practices leads to a problem rather different from the existing straightforward practice that follows a rotation cycle of artificial regeneration, thinning of inferior trees and a clearcut. A lack of realistic models and methods suggesting how to manage uneven-aged stands in a way that is economically viable and ecologically sustainable creates difficulties in adopting this new management practice. To tackle this problem, we make a two-fold contribution in this paper. The first contribution is the proposal of an algorithm that is able to handle a realistic uneven-aged stand management model that is otherwise computationally tedious and intractable. The model considered in this paper is an empirically estimated size-structured ecological model for uneven-aged spruce forests. The second contribution is on the sensitivity analysis of the forest model with respect to a number of important parameters. The analysis provides us an insight into the behavior of the uneven-aged forest model.

Go to link
Read More

Journal Articles | 2017

Approximated set-valued mapping approach for handling multiobjective bilevel problems

Ankur Sinha, Pekka Malo, and Kalyanmoy Deb

Computers & Operations Research

A significant amount of research has been done on bilevel optimization problems both in the realm of classical and evolutionary optimization. However, the multiobjective extensions of bilevel programming have received relatively little attention from researchers in both the domains. The existing algorithms are mostly brute-force nested strategies, and therefore computationally demanding. In this paper, we develop insights into multiobjective bilevel optimization through theoretical progress made in the direction of parametric multiobjective programming. We introduce an approximated set-valued mapping procedure that would be helpful in the development of efficient evolutionary approaches for solving these problems. The utility of the procedure has been emphasized by incorporating it in a hierarchical evolutionary framework and assessing the improvements. Test problems with varying levels of complexity have been used in the experiments.

Go to link
Read More

Journal Articles | 2017

Evolutionary algorithm for bilevel optimization using approximations of the lower level optimal solution mapping.

Ankur Sinha, Pekka Malo, and Kalyanmoy Deb

European Journal of Operational Research

Bilevel optimization problems are a class of challenging optimization problems, which contain two levels of optimization tasks. In these problems, the optimal solutions to the lower level problem become possible feasible candidates to the upper level problem. Such a requirement makes the optimization problem difficult to solve, and has kept the researchers busy towards devising methodologies, which can efficiently handle the problem. Despite the efforts, there hardly exists any effective methodology, which is capable of handling a complex bilevel problem. In this paper, we introduce bilevel evolutionary algorithm based on quadratic approximations (BLEAQ) of optimal lower level variables with respect to the upper level variables. The approach is capable of handling bilevel problems with different kinds of complexities in relatively smaller number of function evaluations. Ideas from classical optimization have been hybridized with evolutionary methods to generate an efficient optimization algorithm for a wide class of bilevel problems. The performance of the algorithm has been evaluated on two sets of test problems. The first set is a recently proposed SMD test set, which contains problems with controllable complexities, and the second set contains standard test problems collected from the literature. The proposed method has been compared against three benchmarks, and the performance gain is observed to be significant. The codes related to the paper may be accessed from the website http://bilevel.org.

Go to link
Read More

Journal Articles | 2017

A novel sandwich algorithm for empirical Bayes analysis of rank data

Arnab Kumar Laha, Somak Dutta, and Vivekananda Roy

Statistics and its interface

Rank data arises frequently in marketing, finance, organizational behavior, and psychology. Most analysis of rank data reported in the literature assumes the presence of one or more variables (sometimes latent) based on whose values the items are ranked. In this paper we analyze rank data using a purely probabilistic model where the observed ranks are assumed to be perturbed versions of the true rank and each perturbation has a specific probability of occurring. We consider the general case when covariate information is present and has an impact on the rankings. An empirical Bayes approach is taken for estimating the model parameters. The Gibbs sampler is shown to converge very slowly to the target posterior distribution and we show that some of the widely used empirical convergence diagnostic tools may fail to detect this lack of convergence. We propose a novel, fast mixing sandwich algorithm for exploring the posterior distribution. An EM algorithm based on Markov chain Monte Carlo (MCMC) sampling is developed for estimating prior hyperparameters. A real life rank data set is analyzed using the methods developed in the paper. The results obtained indicate the usefulness of these methods in analyzing rank data with covariate information.

Go to link
Read More

Journal Articles | 2017

Distribution of Traffic Accident Times in India - Some Insights using Circular Data Analysis

Arnab Kumar Laha, Pravida Raja A.C., and Dilip Kumar Ghosh

International Journal of Business Analytics and Intelligence

Traffic accidents are a major hazard for travellers on Indian roads. These are caused by a variety of reasons including the bad condition of roads, traffic density, lack of proper training of drivers, slack in enforcement of traffic rules, poor road lighting etc. It is further known that certain times of the day are more prone to traffic accidents than others. In this paper we investigate the distribution of traffic accident times using the data published annually by the National Crime Records Bureau (NCRB) over the period 2001-2014 using the tools of circular data analysis. It is seen that the observed distribution of the traffic accident times in most years is bimodal. Thus, several modelling strategies for bimodal distributions are tried which include fitting of mixture of von-Mises distributions and mixture of Kato-Jones distribution. It is seen from this analysis that the distribution of the traffic accident times are changing over the years. Notably, the proportion of accidents happening in late night has reduced over the years while the same has increased for late evening hours. Some more insights obtained from this analysis are also discussed.

Go to link
Read More

Journal Articles | 2017

Through the looking Glass: Role of construal level on description-intensive reviews

Swagato Chatterjee and Aruna Divya T

Advances in Consumer Research

Focus on consumer engagement has led service providers to explore contextual factors influencing consumers’ satisfaction. In this paper, we draw insights from Construal Level Theory to identify the conditions when own vs. others’ experiences along with Process vs. Outcome attributes of services become more important in overall service evaluation

Go to link
Read More

Journal Articles | 2017

Please do interrupt, but nicely! The effect of positive and negative interruptions on product evaluation and choice

Ankur Kapoor and Arvind Sahay

Advances in Consumer Research, 45, 701-702

This research studies the affective consequences of interruptions on evaluation and choice. Six studies demonstrate that positive (negative) interruptions lead to unfavourable (favourable) evaluation and lower (higher) choice of pre-interruption products; but favourable (unfavourable) evaluation and higher (lower) choice of post-interruption products. Relevant mediation and moderation effects are also found.

Go to link
Read More

Journal Articles | 2017

Do celebrities have it all? Context collapse and the networked publics

Asha Kaul and Vidhi Chaudhri

Journal of Human Values

With the advent of social media and increase in networked publics, context collapse has emerged as a critical topic in the discussion of imagined audiences and blurring of the private and the public. The meshing of social contexts portends problematic issues as messages inadvertently reach unimagined audiences causing shame and leading to loss of ‘face’. In this article, we specifically study the impact of context collapse on some celebrities ‘who had it all’ yet, lost ‘it some’ to the world of networked public. The article examines celebrities sharing identity information across multiple contexts and explores situations of lost fame when ‘face’ is threatened, usage falters and breaks some of the well-established norms of interactivity. It concludes that lack of prudence in separating social contexts, loss of ‘face’ and social approval can dampen online celebrity presence. It proposes the use of ‘polysemy’ to simultaneously appeal to audiences from different contexts.

Go to link
Read More

Journal Articles | 2017

Automobile dependence and physical inactivity: Insights from the California Household Travel Survey

Saikat Chakraborty and Eun Jin Shin

Journal of Transport and Health

Background

Auto-dependence has been linked to the physical inactivity epidemic across U.S. cities, resulting in unprecedented increases in incidences of obesity, cardiovascular diseases, depression, etc. The search for strategies to pull an overwhelming majority of auto-dependents out of their sedentary lifestyles by encouraging them to use transit, walk and bike continues to challenge planners and policy-makers.

Methods

We use the 2012–13 California Household Travel Survey data for analyzing the auto-dependence and physical inactivity connection. We select a sample of employed individuals with access to car in urban California, and classify them as discretionary transit riders (N=390), active auto-dependents (N=1287), or sedentary auto-dependents (N=8754) based on their self-reported travel mode use and time spent in physical activity over a 24-h period. We investigate factors that are associated with significantly high physical activity among some auto-dependents relative to the sedentary majority. We also revisit the transit-physical activity connection, and explore conditions that make transit use unfeasible for some active auto-dependents.

Results

Discretionary transit use is associated with higher physical activity. However, there is large variation in physical activity within auto-dependents; significantly higher physical activity is associated with factors such as higher income, flexible work schedule, shorter work hours, and mixed land use. Kids, inflexibility of work schedule, low residential density, lack of pedestrian and bicycling friendly street design, and long distance to transit stops prohibit otherwise active auto-dependents from choosing transit. Employment sector influences both physical activity and choice of transit.

Conclusion

To get sedentary auto-dependents out of endemic physical inactivity, our research indicates the need for targeting lower-incomes, incentivizing employers to provide flexible work hours, and to continue dense, mixed-use developments that make active travel feasible. In addition, to get active auto-dependents to use transit, transit managers must focus on retaining immigrant riders and non-Hispanic Asians, and attracting people with children.

Go to link
Read More
IIMA