No description available.

No description available.

No description available.

The modern trend in electricity industries around the world is towards privatisation. Increased competition, it is argued, will ultimately benefit the consumer. However, the particular nature of electricity generation and supply means strong regulation of a privatised market will always be necessary. In establishing a privatised industry, decisions need to be made about the mechanisms governing the requirements to meet demand, to maintain the viability of the network and to ensure generators are paid correctly for power generated. Unfortunately, it is unclear what processes to use to achieve these goals while still delivering some benefit to the consumer in the form of reduced electricity costs. This research, sponsored by the National Grid Company, examines whether the application of new ideas in artificial intelligence could offer the potential for gaining insights into the affects of certain market mechanisms on the competitors in the market. Our approach to gaining greater understanding into how the market operates is to adopt an evolutionary economics perspective. We have constructed autonomous adaptive agents to represent the generating companies in a simplified model of the UK market in electricity generation. The main body of the thesis contains a description of the process of developing the model and the agent architecture. Once we were satisfied that the model incorporated some key features of the real world market and that the agents, based on learning classifier systems, were able to perform well in simpler environments, we examined how multiple adaptive agents learn to interact in the simplified model. We conclude that the agents are able to learn how to behave in ways analogous to the observed behaviour of real world generating companies. We then illustrate the potential for this type of economic model by examining how alterations to market structure affect agent behaviour, and investigate to what extent the agents are able to learn how to cooperate for mutual long term benefit.

Utility coordination is an exceedingly complex effort of managing, communicating, and facilitating the avoidance and relocation of utility facilities as needed for highway projects. Utility coordination occurs throughout the design and delivery of a project and best practices are used to make sure this occurs efficiently and in the best interest of the public, who are not only the taxpayers but also the ratepayers. Recent research has attempted to enhance utility location technology and procedures, instill frameworks and tools for utility coordination, and proceduralize risk management relative to utility coordination. However, research attempting to improve various aspects of utility coordination simultaneously has led to a lack of consensus on how to integrate these research efforts into an effective standard of practice. These is also not a standard of practice for quantifying utility related risks for transportation projects. This research will attempt to build consensus and contribute to the body of knowledge in this area of utility coordination by presenting an approach to assess the relative utility risks of a project and align current and new practices to minimize those risks. Through statistical analysis of historical project data regarding utility coordination schedules and costs for transportation projects in Kentucky, this study was able to produce a model that estimates utility related risk early in transportation project development. With input and evaluation by subject matter experts, utility coordination best practices were collected and aligned to utility risks on transportation projects. A decision support tool was developed to assist in the use of the mathematical utility risk model and the best practices associated with the varying risk levels. This research also finds that there are disparities among utility stakeholders on transportation projects in regard to the effectiveness or satisfaction with particular best practices. This finding presents the need for early involvement and collaborative utility coordination to select practices that ensure utility related issues on transportation projects are minimized. The research also presents that increased use of alternative contracting methods can pose significant challenges to utility coordination on transportation projects. This stems from the finding that utility coordination practices were not uniformly effective across these varying procurement methods. Furthermore, as Departments of Transportation continue to deal with resource issues, one of which being manpower within utility coordination, the use of consultants for utility coordination presents its own set of complexities. The research finds the best application of consult-led utility coordination is through third-part consultants specializing in utility coordination, those who have been state-specifically trained for utility coordination, and prequalified for utility coordination work.

In this thesis, we present a decision-theoretic framework for building decision support systems that incrementally elicit preferences of individual users over multiattribute outcomes and then provide recommendations based on the acquired preference information. By combining decision-theoretically sound modeling with effective computational techniques and certain user-centric considerations, we demonstrate the feasibility and potential of practical autonomous preference elicitation and recommendation systems. More concretely, we focus on decision scenarios in which a user can obtain any outcome from a finite set of available outcomes. The outcome is space is multiattribute; each outcome can be viewed as an instantiation of a set of attributes with finite domains. The user has preferences over outcomes that can be represented by a utility function. We assume that user preferences are generalized additively independent (GAI), and, therefore, can be represented by a GAI utility function. GAI utilities provide a flexible representation framework for structured preferences over multiattribute outcomes; they are less restrictive and, therefore, more widely applicable than additive utilities. In many decision scenarios with large and complex decision spaces (such as making travel plans or choosing an apartment to rent from thousands of available options), selecting the optimal decision can require a lot of time and effort on the part of the user. Since obtaining the user's complete utility function is generally infeasible, the decision support system has to support recommendation with partial preference information. We provide solutions for effective elicitation of GAI utilities in situations where a probabilistic prior about the user's utility function is available, and in situations where the system's uncertainty about user utilities is represented by maintaining a set of feasible user utilities. In the first case, we use Bayesian criteria for decision making and query selection. In the second case, recommendations (and query strategies) are based on the robust minimax regret criterion which recommends the outcome with the smallest maximum regret (with respect to all adversarial instantiations of feasible utility functions). Our proposed framework is implemented in the UTPref recommendation system that searches multiattribute product databases using the minimax regret criterion. UTPref is tested with a study involving 40 users interacting with the system. The study measures the effectiveness of regret-based elicitation, evaluates user comprehension and acceptance of minimax regret, and assesses the relative difficulty of different query types.

In this thesis, we present a decision-theoretic framework for building decision support systems that incrementally elicit preferences of individual users over multiattribute outcomes and then provide recommendations based on the acquired preference information. By combining decision-theoretically sound modeling with effective computational techniques and certain user-centric considerations, we demonstrate the feasibility and potential of practical autonomous preference elicitation and recommendation systems. More concretely, we focus on decision scenarios in which a user can obtain any outcome from a finite set of available outcomes. The outcome is space is multiattribute; each outcome can be viewed as an instantiation of a set of attributes with finite domains. The user has preferences over outcomes that can be represented by a utility function. We assume that user preferences are generalized additively independent (GAI), and, therefore, can be represented by a GAI utility function. GAI utilities provide a flexible representation framework for structured preferences over multiattribute outcomes; they are less restrictive and, therefore, more widely applicable than additive utilities. In many decision scenarios with large and complex decision spaces (such as making travel plans or choosing an apartment to rent from thousands of available options), selecting the optimal decision can require a lot of time and effort on the part of the user. Since obtaining the user's complete utility function is generally infeasible, the decision support system has to support recommendation with partial preference information. We provide solutions for effective elicitation of GAI utilities in situations where a probabilistic prior about the user's utility function is available, and in situations where the system's uncertainty about user utilities is represented by maintaining a set of feasible user utilities. In the first case, we use Bayesian criteria for decision making and query selection. In the second case, recommendations (and query strategies) are based on the robust minimax regret criterion which recommends the outcome with the smallest maximum regret (with respect to all adversarial instantiations of feasible utility functions). Our proposed framework is implemented in the UTPref recommendation system that searches multiattribute product databases using the minimax regret criterion. UTPref is tested with a study involving 40 users interacting with the system. The study measures the effectiveness of regret-based elicitation, evaluates user comprehension and acceptance of minimax regret, and assesses the relative difficulty of different query types.

In activity based modelling the concept of rescheduling is very important in order to gain dynamic scheduling of activities and to adjust the effect of unexpected incidents in individual agendas to keep them realistic and valid. This report describes a new framework to investigate algorithms for rescheduling on a large scale. This framework models the information of traffic demand and results of micro simulation of traffic on a loaded network; it enables agents to adapt their schedules by providing them with information about the traffic flow. A perception filter for each agent is included in this framework. It models the concept that some agents can notice the broadcast traffic information about the incident and get their own prediction of the expected delay, while other agents who do not notice the information can become aware only by experiencing traffic jam. Initial agendas are created by means of the FEATHERS activity based schedule generator for mutually independent agents. FEATHERS has no knowledge about the actual transportation network but makes use of an impedance matrix that specifies the minimal travel time between traffic analysis zones. The matrix specifies a free-flow value for the uncongested case and correction values for the loaded network. In this new framework the network state can be changed by agent behaviour and external incidents; the effect of this change in network state is perceived differently by each agent through a perception filter, and according to the perceived value individual adaption is calculated by a ReScheduler. The modified behaviour again creates new traffic demand hence creating a new traffic state; this phenomenon continues for the complete day. Each activity in the agenda is assumed to generate some utility. Each individual is assumed to maximize the total utility over the day. The ReScheduler is implemented using a marginal utility function that monotonically decreases with activity duration. This results in a monotonically converging relaxation algorithm to efficiently determine the new activity timing when less time is available for activities due to increased travel time caused by the incident.

No description available.

Construction projects are complicated and fraught with so many risks that many projects are unable to meet pre-defined project objectives. Managers of construction projects require decision support tools that can be used to identify, analyze and implement measures that can mitigate the effects of project risks. Several risk analysis techniques have been developed over the years to enable construction project managers to make useful decisions that can improve the chances of project success. These risk analysis techniques however fail to simultaneously address risks relating to cost, schedule and quality. Also, construction projects may have scarce resources and construction managers still bear the responsibility of ensuring that project goals are met. Certain projects require trade-offs between technical and managerial risks and managers need tools that can help them do this. This thesis evaluates the usefulness of the Advanced Programmatic Risk Analysis and Management Model (APRAM) as a decision support tool for managing construction projects. The development of a visitor center in Midland, Texas was used as a case study for this research. The case study involved the implementation of APRAM during the concept phase of project development to determine the best construction system that can minimize the expected cost of failure. A risk analysis performed using a more standard approach yielded an expected cost of failure that is almost eight times the expected cost of failure yielded by APRAM. This study concludes that APRAM is a risk analysis technique that can minimize the expected costs of failure by integrating project risks of time, budget and quality through the allocation of resources. APRAM can also be useful for making construction management decisions. All identified component or material configurations for each alternative system however, should be analyzed instead of analyzing only the lowest cost alternative for each system as proposed by the original APRAM model. In addition, it is not possible to use decision trees to determine the optimal allocation of management reserves that would mitigate managerial problems during construction projects. Furthermore, APRAM does not address the issue of safety during construction and assumes all identifiable risks can be handled with money.