Remica Aggarwal

Bayesian network on labour dissonance: a social sector development challenge to India

Indias unorganized labour force contributes about one third of the total labour sector. The scenario is even worse in the Indian automotive industry which employs a little over 7% on a permanent basis. Problems get exaggerated due to the outdated labour laws, ironically established to support and protect workers. The disappointing areas in the labour contract act and labour laws have led to unfair wage practices and a hostile work environment, giving way to labour discord. This research paper discusses the key issues of labour dissonance in the Indian automobile industry using a Bayesian network analysis. Real-life case-study examples from the Indian automobile industry were considered to identify the rationale behind labour unrest. Bayesian analysis of a set of 250 responses helped us to understand the associations among key attributes of labour dissatisfaction.

Optimal duration of advertising campaigns for successive technology generations using innovation diffusion theory

Global market and tough competition compels a firm to continuously conceive new ideas and introduce new technologies in the market. As a result, often more than one generation products compete in the same market; creating an incredible pressure on managers for balanced advertising campaigns for the existing product generations. Advertising of multi-generation product involves selection of appropriate advertising medium, analysing the target market and appropriate utilisation of the available advertising budget. Effective advertising campaign is critical for success of a product in the market. Hence, finding the optimal advertising campaign duration is important as huge chunk of a firms budget is allocated for this purpose. For, successive technology generations, advertising at right time become even more important. This study developed a mathematical model to determine the optimal duration of advertising campaigns for successive generations product based on diffusion of information in a social group. The optimal timing depends on diffusion coefficient, population size, advertising cost per time unit, unit price, etc.

Non Dominated Sorting Genetic Algorithm for Chance Constrained Supplier Selection Model with Volume Discounts

This paper proposes a Stochastic Chance-Constrained Programming Model (SCCPM) for the supplier selection problem to select best suppliers offering incremental volume discounts in a conflicting multi-objective scenario and under the event of uncertainty. A Fast Non-dominated Sorting Genetic Algorithm (NSGA-II), a variant of GA, adept at solving Multi Objective Optimization, is used to obtain the Pareto optimal solution set for its deterministic equivalent. Our results show that the proposed genetic algorithm solution methodology can solve the problems quite efficiently in minimal computational time. The experiments demonstrated that the genetic algorithm and uncertain models could be a promising way to address problems in businesses where there is uncertainty such as the supplier selection problem.

Optimal Advertisement Planning for Multi Products Incorporating Segment Specific and Spectrum Effect of Different Medias

A large part of any firm’s investment goes in advertising and therefore planning of an appropriate media for advertisement is the need of today so as to achieve the best returns in terms of wider reach over potential market. In this paper, we deal with a media planning problem for multiple products of a firm in a market which is segmented geographically into various regional segments with diverse language and cultural base. As such each of these regional segments responds to regional advertising as well as national advertising which reaches them with a fixed spectrum. The objective is to plan an advertising media (national and regional media) for multiple products in such a way that maximizes the total reach which is measured through each media exclusively as well as through their combined impact. The problem is formulated as a multi-objective programming problem and solved through goal programming technique. A real life case is provided to illustrate the applicability of the proposed model.

Identifying and Prioritizing Human Capital Measurement Indicators for Personnel Selection Using Fuzzy MADM

Human capital constitutes an important and essential asset in any organization. Identifying and evaluating various human capital indicators or assessors is a multi-attribute decision making problem (MADM) which includes both qualitative and quantitative factors and therefore an appropriate MADM technique is required for proper evaluation and assessment of these factors or attributes. This study aims at defining a methodology based on Delphi method as well as Fuzzy Analytic Hierarchy Process (FAHP) is proposed to prioritize various human capital indicators associated with the five main attributes i.e. Talent, Integration, enabling a performance-based culture/climate, capability and leadership. The findings showed that in context with the case problem, Employees satisfaction with advancement opportunities, Internal relationship index, Employee skills, Creating results by using knowledge, Percentage of employees with access to appropriate training and development opportunities are the most important indicators for the HC in an Indian organization.