Papers by Gianluca Murgia
IMA Journal of Management Mathematics, 2015
ABSTRACT Multi-stage stochastic programming can support large consumers in developing electricity... more ABSTRACT Multi-stage stochastic programming can support large consumers in developing electricity portfolios that balance the expected total cost and the risk level. Nevertheless, the adoption of multi-stage stochastic programming in real-world problems is often made difficult by the high computational burden required. In this paper, we present an innovative approach, called General Policy Function Approximation, that provides good solutions to the electricity portfolio problem in a limited computational time, owing to the integration of multi-stage stochastic programming and machine learning. Our approach improves the Policy Function Approximation (PFA) approach proposed by Defourny et al. [(2012). Multi-stage stochastic programming: A scenario tree-based approach to planning under uncertainty. Decision Theory Models for Applications in Artificial Intelligence (L. E. Sucar, E. Morales, F. Eduardo & J. Hoey eds). vol. 6. Hershey: IGI Global, pp. 97–144], by developing a single policy function generated from a larger amount of data. Owing to a realistic computational campaign, we show that our approach outperforms PFA both in terms of quality of the policy obtained, and in terms of time required.
Methods for Decision Making in an Uncertain Environment, 2012
"The current financial and economic crisis has strengthened the importance of correctly scor... more "The current financial and economic crisis has strengthened the importance of correctly scoring and managing distressed debts. In scientific literature, the scoring of distressed debts has been handled using different methods, while the interest in managing those debts is limited to the choice of the best collection activity. We develop a Decision Support System which scores the recovery rate of each distressed debt, starting only from a limited set of debt’s features, and calculates the debts’ daily planning and their assignment to the collection agency operator. Our DSS, which integrates artificial neural network, Analytic Hierarchy Process, integer programming, and hidden Markov model, is already under experimentation. Now, we have validated the scoring of the recovery rate of the packages of debts, comparing our results with those obtained by a collection agency, and we achieved a classification performance very similar to other methods presented in literature. Besides, we compared also our scoring system with logistic regression, Bayesian classifier, and regression tree, supporting the primacy of artificial neural network on these methods, in accordance with the conclusions of previous literature. "
Proceedings of the IADIS International Conference on Cognition and Exploratory Learning in the Digital Age, CELDA 2010, 2010
In this paper we analyze an e-learning course for managerial education over three years making us... more In this paper we analyze an e-learning course for managerial education over three years making use of the social network analysis. Our aim is to represent the knowledge flows within a virtual learning environment and to identify some keyroles, previously highlighted by Cross and Prusak (2002) in organizational contexts. Our methodology shows the specific contribution brought by each actor to the knowledge development in a community, so it can support teachers in the development of their teaching strategies. © 2010 IADIS.
The characteristics of several issues, such as ICT infrastructure, learning objects, multimedia e... more The characteristics of several issues, such as ICT infrastructure, learning objects, multimedia effects, way of teaching and tutoring, and interaction among the participants, all affect the quality of an e-learning system, though their impact varies accordingly with the specific e-learning system analysed. The object of our analysis is the e-learning system used in the on-line courses offered by the School of Engineering in the University of Rome "Tor Vergata". The success of this system is mainly based on a deep level of interaction among instructors and students, mostly through the use of on-line forums, which allow both discussions related to the learning process (including class notes, homework, and assessments) as well as favour the social cohesion of the classrooms, thus improving the students' own rate of learning. In this paper, we present the results of a DSS prototype that focus on such aspects of on-line interaction and discuss the lessons learned from its first development.
This work deals with the development of a business game, named Board Leadership Game (BLG), which... more This work deals with the development of a business game, named Board Leadership Game (BLG), which is still under further development at the University of Rome "Tor Vergata". The final goal is to build a powerful didactic tool for students and management professionals in order to favor the understanding of underlying dynamics in the decisions' formulation processes inside groups; this also takes into account principles that come from the Upper Echelons Theory, developed in management science, which establishes a direct relationship between top management's demographic characteristics and the strategies carried out by the firm. In the developed framework, each player plays the company CEO, making up strategic proposals that are examined by the company's Board of Directors (BoD) and could be eventually changed in accordance with the top management demographic characteristics and with CEO's leadership (the ability to influence the BoD's final decisions). Strategies chosen by the BoD, together with market's characteristics and competitors' actions, determine the company's economic and financial results, which in turn close the loop by having a direct influence on CEO's leadership.
Uploads
Papers by Gianluca Murgia