<p>The human computer interface is a vital component of the computer system, and is often seen as the entire system from the user's perspective. A properly designed interface should allow the user to effortlessly access information, which can also facilitate the decision making process. In this study, we develop a research framework to explore specific links between the decision making, information access and user interface domains. We examine three specific interface Wide Web history mechanisms) and their links within our research framework, which can then provide design implications for usable interfaces. Voice and text output modes were investigated in a multiple alternative decision making task experiment. We were able to conduct a direct comparison between these output modes by presenting the same information for the same length of time for each mode (voice, text, and both voice and text). We found the text mode was the most efficient, resulting in shorter decision times. Both voice and text was preferred to voice alone, but there were no significant differences between text\voice and both\text output mode pairings. A multiple alternative decision making task experiment was also used information structure was not imposed on the user. We found that a top-down search strategy was directly related to a tendency to use a compensatory and alternative-wise decision making strategy. We also found that the propensity to use a top-down strategy became less as the search progressed in favour of a more opportunistic approach. The final interface component investigated was a World Wide Web hypermedia history mechanism. A Memory Extender Mechanism for Online Searching (MEMOS) tool was developed to provide the Web user with intra- and inter-sessional navigation support in parallel with existing browser navigation mechanisms. This tool can be used to navigate among pages within a particular session, or save all or part of a navigation session for future inter-sessional use. Session sites may be saved in a three level user-defined hierarchy, thus making lists more manageable and understandable. We conducted experiments where subjects compared MEMOS with corresponding Netscape 3.0 Web browser history mechanisms. The results of the experiments indicated significant preferences for the MEMOS tool, and significant efficiency and effectiveness measures for MEMOS inter-sessional was also shown to be effective. From our examination of the above interface components we established specific interactive relationships among the derision making, information access and user interface domains. These relationships contribute to a better understanding of interface components that can be used to develop usable systems. Our framework also suggests possible directions for future research.</p> / Doctor of Philosophy (PhD)

<p>Selecting projects to develop from the many that are usually possible, or "project portfolio selection " is a crucial decision in many organizations. Many different methods for portfolio selection have been suggested in the literature but no integrated framework has been developed for carrying it out. In this thesis, we suggest a framework that builds on the strengths of existing methods. The proposed framework separates the project portfolio analysis and selection process into distinct stages and allows users to choose the techniques they find the most suitable for each stage. Since finding the optimal solution is the most sophisticated part of the proposed framework, we developed a zero-one linear integer-programming model for this stage that overcomes the shortcomings of existing models and advances previous work in this area. The model we developed considers multiple, conflicting goals, including qualitative objectives, and selects and schedules the optimal set of projects. The solution maximizes benefits according to pre-specified priorities without violating any constraints such as resource limitations, project interdependencies, and portfolio balancing. We implemented the proposed framework in the form of a decision support system (DSS) to facilitate and encourage decision maker involvement throughout the process. For this purpose we developed PASS (Project Analysis and Selection System). PASS is a user-friendly DSS that suits the knowledge level of potential users and allows interaction with the system as well as intervention to make adjustments to the solution it provides. We tested the usefulness, perceived usefulness and perceived ease of use of PASS in a lab setting and gained significant results in support of the developed hypotheses. We also presented our approach to two high-tech firms who wanted to investigate the feasibility of PASS for project portfolio selection, and received very promising feedback from these companies.</p> / Doctor of Philosophy (PhD)

<p>This thesis describes efficient algorithms and complexity results for some machine scheduling and related problems, which are encountered in automated manufacturing systems. We introduce a new class of robotic-cell scheduling models. The novel aspect is that parts need to reenter machines several times before they are finished. The problem is to find the sequence of robot move cycles and the part processing sequence that jointly minimize the cycle time or the makespan. We show that the problems are computationally intractable with three machines and present polynomial solutions for a variety of two-machine configurations. We then consider the problem of scheduling multi-component parts in a two-machine robotic cell, where each part is composed of K identical components to be processed together on the first machine, then processed on the second machine individually. We study the cycle time and makespan minimization problems, and show that both are polynomially solvable. We investigate the problem of minimizing cycle time in a two-machine job shop, where each job has at most three operations. We reduce the problem to a two-machine reentrant flow shop problem. By extending previous results on the reentrant flow shop problem, we propose a new pseudo-polynomial algorithm, as well as a fully polynomial-time approximation scheme for certain special cases of the job shop problem. We also describe a 4/3-approximation algorithm for the general problem, and identify several well-solvable cases. Finally, we study special cases of the traveling salesman problem on permuted Monge matrices, which arose from robotic-cell scheduling problems. By using the theory of subtour patching, we reduce the problems to finding a minimum-b -weight spanning tree in the patching graph. In general, this problem is [Special characters omitted.] -hard. We show, however, that newly defined special properties of the distance matrix allow us to find in polynomial time a minimum-b -weight spanning tree, and thus an optimal tour, for these new classes.</p> / Doctor of Philosophy (PhD)

<p>This dissertation presents a collection of research projects on software agents in electronic commerce (e-commerce). A common theme throughout this research is that agents are an innovation. We are interested in identifying conditions and design criteria that would lead to their adoption in e-commerce applications. We define and study a class of agent applications that fall under the Decision Support Systems (DSS) Approach, where users delegate part of a decision-making task to a software agent. We show how findings from traditional DSS research can guide the development of e-commerce applications that include software agents. Two frameworks are presented that organize research and development activity. The first framework looks at the kinds of knowledge that agents should possess if they are to assist in e-commerce decision-making and identifies some of the major research challenges in designing intelligent agent applications. The second framework is directed at development and design activities. It builds on models of buyer behaviour where perceived risk and frequency of purchase are two characteristics of purchasing situations that can help identify when buyers are expected to find agents useful. The results of two empirical studies suggested by these frameworks are presented. The first study was exploratory and identified consumer preferences for information display over searching and browsing tasks. This was a first step in a larger project aimed at designing adaptable agents to support consumers in different information-seeking modes. A second experiment studied consumer behaviour in the actual online purchase of a music compact disk. We found that subjects who used an agent made their purchase decisions in less time and made more-informed decisions than subjects who did not use an agent.</p> / Doctor of Philosophy (PhD)

<p>This thesis develops quality control and improvement techniques based on grouped data. Grouped data occur frequently in industry. However, in the past, most techniques have failed to directly take this grouping into account, and as a result do not perform well in many circumstances. Two major areas of application are considered. First, acceptance sampling plans, acceptance control charts, and Shewhart control charts based on grouped data are developed. These forms of statistical process control have broad application and are in use widely. The design and implementation methodology is derived assuming either a normal or Weibull process, but is easily adapted to any other underlying distribution. A number of design approaches are presented and their relative advantages and disadvantages are discussed. The second application involves estimating the correlation between destructively measured strength properties. This problem arises in the area of structural design. To obtain an estimate of the correlation censoring of the strength data is required. The censoring or proof-testing results in grouped data. A number of simple estimation procedures are presented and compared.</p> / Doctor of Philosophy (PhD)

<p>This thesis deals with game theory and its applications in management science and focuses upon some management science areas such as inventory control and new product development. Some interesting theoretical findings and new policies are obtained by using the game theoretical approach to analyze certain management science problems. The discussion starts with a review of static game theory models and their applications in management science. Of particular interest here is the state of the art of game theory as an analytical technique in management science. Chapters Two and Three discuss the discount problem. Particular attention is paid to the gaming nature and the buyer's demand aspect of the problem. It is shown that, if they work independently and rationally, the seller and the buyer can gain from price discount only if it can attract more demand from the buyer. Nevertheless, they can gain from quantity discount even if demand is constant. Quantity discount is always better than a price discount for the seller and, in certain situations, can be very efficient in obtaining the maximum profit. Optimal decisions are obtained for both the seller and the buyer under various conditions. Chapter Four studies the order quantities of substitutable products with stochastic demands. This analysis extends the newsboy problem analysis into situations with three or more players. It is shown that there is one Nash equilibrium for the problem. If any player(s) acts irrationally, the other players' decision problem reduces to the one without the irrational player(s). If cooperation is possible, their decisions depend on whether side payments are allowed. If side payments are allowed, they will determine their order quantities together. If side payments are not allowed, secure strategies exist for each player. It is also shown that all players' cooperation is often worthwhile and feasible. Chapter Five analyzes the growth of new repeat purchasing products. It is shown that markets of repeat purchasing products will never saturate like that of consumer durable products unless customers are extremely loyal to at least one product. For new repeat purchasing products, the optimal advertising strategy is increasing at the introductory stage and then decreasing or possibly terminating after some time and, the optimal service strategy is monotonically increasing at the introductory stage and then possibly maintained constant at a certain level. Especially, more advertising should be done at early stages against competitors' advertising campaign. The game is solved analytically for optimal strategies in the case where all the control functions representing the effects of advertising and service are linear in the control variables. Finally, the main findings and possible extensions to this research are briefly summarized in Chapter Six. (Abstract shortened by UMI.)</p> / Doctor of Philosophy (PhD)

<p>This thesis represents research in the combined areas of inventory and maintenance. It analyzes two independent inventory and maintenance problems under dynamic systems: (i) a production and maintenance problem and (ii) a repairable-item inventory problem. For each problem, the thesis develops a new control model and proposes a simultaneous determination of optimal inventory and maintenance policies. The first part of the thesis examines a production process where the process performance deteriorates over time in the absence of preventive maintenance. First, it develops a new finite-time control model for optimal production and maintenance decisions by combining a dynamic maintenance model with a production control model. Second, it derives the necessary conditions for optimal production and maintenance controls using the maximum principle. Finally, it proposes two optimization algorithms for numerically solving the necessary conditions already derived. The second part of the thesis considers the repairable-item inventory problem, which may be faced at each period by the inventory manager responsible for determining the optimum quantities to purchase new serviceable units, to repair and to junk returned repairable units in order to satisfy random demand for serviceable units. First, it proposes an inventory model for repairables, incorporating several important features. The model includes a periodic review policy, random demand, lost sales for unsatisfied demand, set-up costs for ordering and repair, and a dynamic return process. Second, it employs a quite different solution methodology from what the previous research has used. The approach employed here is a 'Markov decision process (MDP)'. With this approach, the inventory problem is remodelled as a discrete-time Markov decision problem with two-dimensional state and three-dimensional decision spaces and then solved for finite-time planning horizon using the backward induction algorithm and for infinite-time planning horizon using the method of successive approximations. Finally, it introduces and utilizes two acceleration techniques, the error bounds approach and State Decomposition by Dimension (SDD), for speeding up the convergence of the computational methods described above.</p> / Doctor of Philosophy (PhD)

<p>This study introduces a new model for the planar maximal covering location problem (PMCLP) under different block norms. The problem involves locating p facilities anywhere on the plane in order to cover the maximum number of n given demand points. The generalization we introduce is that distance measures assigned to facilities are block norms of different types and different proximity measures. This problem is handled in three phases. First, a simple model based on the geometrical properties of the block norms' unit ball contours is formulated as a mixed integer program (MIP). The MIP formulation is more general than previous PMCLP's and can handle facilities with different coverage measures under block norm distance and different setup cost, and capacity. Second, an exact solution approach is presented based on: (1) An exact algorithm that is capable of handling a single facility efficiently. (2) An algorithm for an equivalent graph problem--the maximum clique problem (MCP). Finally, the PMCLP under different block norms is formulated as an equivalent graph problem. This graph problem is then modeled as an unconstrained binary quadratic problem (UQP) and solved by a genetic algorithm. Computational examples are provided for the MIP, the exact algorithm, and the genetic algorithm approaches.</p> / Doctor of Philosophy (PhD)

<p>This study describes algorithms for the solution of several single facility location problems with maximin or minimax objective functions. Interactive computer graphical algorithms are presented for maximizing the minimum rectilinear travel distance and for minimizing the maximum rectilinear travel distance to a number of point demands when there exist several right-angled polygonal barriers to travel. For the special case of unweighted rectilinear distances with barriers, a purely numerical algorithm for the maximin location problem is described. An interactive computer graphical algorithm for maximizing the minimum Euclidean, rectilinear, or general l$\sb{\rm p}$ distance to a number of polygonal areas is described. A modified version of this algorithm for location problems with the objective of minimizing the maximum cost when the costs are non-linear monotonically decreasing functions of distance is presented. Extension of this algorithm to problems involving the minimization of the maximum cost when the costs are functions of both distance and direction is discussed using asymmetric distances.</p> / Doctor of Philosophy (PhD)