Catálogo de publicaciones - libros

The concept of landscape theory provides a way of thinking about the many possible ways in which elements of a system can fit together, predicting which configurations are most likely to occur, how much dissatisfaction with the outcomes is inevitable, and how the system will respond to changes in the relationship between the elements. Recently we extended landscape theory using the notion of fuzzy partitions. In this paper we study the formation of competing alliances by companies who wish to increase their own business efficiency in the civil aviation industry. The purpose of this paper is to evaluate the fuzzy landscape theory by using an airline industry case.

We present an extension to an existing work that investigates conditions for “disruptive technologies” to cause the failure of a former monopolist (incumbent) to the advent of new firm (entrant). This paper shows that 1. the forecast of the positions of the entrant technology allows the incumbent to detect threatening entrant technologies, and 2. cloning the entrant firm allows the incumbent to participate in the new market and assures the survival of the incumbent firm group.

Using an agent-based multi-asset artificial stock market, we simulate the survival dynamics of investors with different risk preferences. It is found that the survivability of investors is closely related to their risk preferences. Among the eight types of investors considered in this paper, only the CRRA investors with RRA coefficients close to one can survive in the long run. Other types of agents are eventually driven out of the market, including the famous CARA agents and agents who base their decision on the capital asset pricing model.

In this paper, using agent-based models, we discuss the effects of Passive Investment Strategies in asset management business. Although the Passive Investment Strategy is an effective way in efficient markets, Behavioral Finance points out that markets aren’t always efficient. We build a virtual financial market which consists of a thousand investors and allows them to trade two types of assets: a stock and a riskless asset. In this market, multiple types of investors exist and conduct trades based on the investment rules defined for each type. The experiments suggest that Passive Investment is valid in a realistic efficient market, yet it could have bad influences such as market instability and inadequate asset pricing deviation.

U-Mart is interdisciplinary education / research project of a virtual futures market on computer for engineering and economics. We have held the international open experiments UMIE 200x which is the competition for machine agents. This paper reports the result of UMIE 2004 and classify the machine agents submitted in UMIE 2002, 2003 2004 in six clssification criterion.

There is a problematic disconnection in much research on organizational behavior between empirical evidence, mostly from case studies, and theorizing, most of which has been discursive rather than formal. This paper presents an empirically based, bottom-up view of organizational behavior, obtained by performing case studies of small firms in the New Jersey plastic injection molding industry, and then developing a multi-agent simulation model of such a firm. The model embodies a formal, explicit, and grounded (limited) theory of organizational behavior. Results suggest that a realistic account of corporate behavior depends in part on representing employees not as homo economicus but as agents having bounded rationality and subject to social influences.

The petrol price market is a highly sensitive and competitive market with many processes combining at different temporal and spatial scales to affect each petrol station’s prices. Previous models developed to represent the relationship between petrol and a variable are empirical and mathematical. These suffer from a number of problems, chiefly: the parameters are all on the same scale (behaviours executed at the ‘micro’ level are not tied to ‘global’ level variables like oil prices); the parameters are often difficult to estimate and lack realism; very little, if any, account of any geographical effects is taken, and, finally, mathematical models by their nature only consider quantitative parameters and therefore miss out on qualitative, behavioural information. The work within this paper presents a series of three multi-agent and hybrid models that seek to rectify some of these problems. The models are behavioural and work at the scale of the individual. The results show that this is a promising method for modelling dynamic, geographical systems.

Conventional supply chain management (SCM) is not always concerned with optimal solutions in terms of product allocation. Virtual market based supply chain operation solves the product allocation problem by distributing the scheduled resources according to the market prices, which define common scale of value across the various products. We formulate supply chain model as a discrete resource allocation problem, and demonstrate the applicability of the Walrasian virtual market concept with multi-agent paradigm to this framework. In this paper we demonstrate the proposed algorithm successfully calculates Pareto optimal solutions in the supply chain product allocation problem by comparing our method with conventional analytic approaches, such as ε constraint method and fixed point algorithm.

Assessing the changing dynamic between the demand that is placed upon a community by cumulative exposure to hazards and the capacity of the community to mitigate or respond to that risk represents a central problem in estimating the community’s resilience to disaster. We present an initial effort to simulate the dynamic between increasing demand and decreasing capacity in an actual disaster response system to determine the fragility of the system, or the point at which the system fails. The results show that access to core information enhances efficiency of response actions and increases coordination throughout the network of responding organizations

An agent simulation test model for pedestrian flow has already been developed, but an experimental simulation that incorporates actual data about an accident has not been designed. In our research, we have reconstructed this accident by improving an existing test model. Based on the accident report and using data gained through spatial research we have revised a pedestrian flow simulator. Through a simulation analysis including rows of standing people and the confrontation of the ascent and descent flows, we have explored measures to prevent a pedestrian accident. The ASPF is to assess measures for managing pedestrian flows by focusing on the domino risk rather than a reconstruction of crowd collapse as in the overpass accident. The simulation results show that a two-way flow, combined with stationary people can trigger an accident even on an overpass that satisfies present design standards. Moreover, we have confirmed that even simple traffic regulations such as partitions can be an effective measure to prevent a pedestrian accident.