Government regulations and legislation, organizational policies, and customer contracts all stipulate rules that an organization and its internal entities must adhere to. Compliance with said rules is often mandatory, and if compliance is not met, may lead to negative consequences. Continuous compliance monitoring involves collecting data on a continuous basis in order to track the relevant business processes to ensure compliance with government regulations and legislation, organizational polices and customer contacts. This thesis presents a framework for continuous compliance monitoring of business to business (B2B) processes in the context of a publish/subscribe architecture for event-driven business process integration. The framework integrates a streaming event data model with an agent-based surveillance portal to provide continuous support for dynamic exception alerts and performance management reporting. The government regulations and legislation, organizational policies and contracts are represented within the framework by event-condition-action (ECA) policies that can be monitored in terms of the event data collected by the surveillance portal. An eHealth scenario in which organizations collaborate to provide regulated at home care is used to illustrate our approach.

Design and operations of electrical distribution-transmission networks are analyzed mathematically, implemented numerically and validated by simulation. A dynamic model of a three node network with capacitors, inductors, load current controllers and regulators is proposed and cast in a general model of differential state-space equations in canonical form. The model is implemented via a Runge-Kutta algorithm. Realistic values of distribution systems are chosen as input and validated interactively so as to avoid instabilities and maintain reasonable characteristics. Typically cases are analyzed and the behavior of state variable is represented graphically. The software used and mode of representation aim at providing a robust environment to help power managers in their daily control of load balancing. The analysis also opens directions for the design of power distribution network.

This paper presents a model for the evaluation of coastal zone sites in conjunction with supporting decision making on the use of potential sites for aquaculture as well as other site activities including commercial fisheries, and as reserves for natural resources. The decision support model captures site specific data in the form of a geographical information system that overlays selected geographical regions with natural resource dynamics, habitat, commercial activities including aquaculture, and influence plumes including toxicology. Descriptive data for selected regions including system overlays and interactions are then evaluated to provide input to a multicriteria analysis that positions decision makers with respect to the relative importance of resources, habitat, commercial activities, and influence plumes. The model compares alternative evaluations of selected regions among diverse users., as well as providing a group decision evaluation procedure to assist in coastal zonal governance decision makers such as the awarding of fish farm site applications. The model is applied to the coastal zone of Grand Manan Island, New Brunswick situated in the Bay of Fundy.

Free/Libre and Open Source Software (FLOSS) has been an important factor in the software world and it has profoundly changed the way people deal with software products. However, it seems that little job appear to have been done to develop digital watermarking systems as FLOSS. Developing such a system requires multiple-disciplinary knowledge of digital image watermarking techniques, digital image processing, information hiding, computer science technology and cryptography. The demands of these diverse subjects may suggest that building such a system is a difficult task. This thesis explores the feasibility of setting up an Internet-based digital image watermarking system capable of embedding and extracting text messages as FLOSS. Several digital watermarking algorithms, such as Least Significant Bit, Patchwork, brightness modulation and DFT amplitude and phase modulations were explored. A choice was made to adopt a J2EE (Java 2 Platform Enterprise Edition) browser/server structure to build the embedding/extraction subsystems and to use Java, HTML/XML and Javascript languages to implement the system, as these are independent of specific computing platforms (eg, Windows or Linux). A trade-off analysis on watermark imperceptibility, robustness, reliability and capacity of algorithms was carried out and taken into consideration in shortening the processing time in the empirical system. The thesis work has developed an illustrative framework for future extension and research.

This thesis discusses a new stream of research and analysis which will form part of the Northern Watch Technology Demonstration project. The objective of the thesis is to develop and illustrate a procedure for identifying and characterizing combinations of sensors and systems that will provide cost-effective options for Arctic maritime surveillance. A ship detection simulation is constructed, the results of which are used to produce a set of ranked options for combinations of sensors used to conduct maritime surveillance at a strategic choke point located at the Barrow Strait, in Canada's Northwest Passage. The overall objective of the surveillance is to improve on the detection, classification, and identification of maritime vessels. The modeled performance and effectiveness of each sensor is evaluated in relation to the multiple objectives using the Analytical Hierarchy Process (AHP) to rank alternative sensor effectiveness and performance. The results indicate that the most cost-effective solution is to install an Automatic Identification System (AIS) sensor at the Northern Watch station. However, due to practical concerns, alternatives to this approach are presented and discussed.

Organizations are frequently expected to improve operational efficiency through effective decision making. Effective decision making, among other things, requires a configurable business process application platform that can be used to execute, monitor and manage processes involving people, teams, departments, organizational units and their multiple applications. This kind of platform facilitates data collection for performance management and analysis for measuring quality of service delivery and business process optimization. It also facilitates collaboration between process participants, regulatory compliance, configurability to fit to changing requirements and accelerates decision making by sensing actionable events. Healthcare is a significant area where such a platform can be of great use for monitoring patient care. Patient care monitoring applications are data intensive and require medical guidelines based collaboration between patients, individuals, care providers and organizational units. These applications should provide performance management information to measure the quality of healthcare service delivery processes. These types of applications must also be configurable to adapt to frequently changing types of business events and the process of collect event data. This thesis proposes a model driven application framework that can be used to create configurable process oriented applications that can be customized to the current needs of an organization. The proposed framework also provides the ability to monitor, manage and report on processes and collected event data. A healthcare scenario is used to illustrate and validate our approach by building a patient monitoring application.

No description available.

No description available.

No description available.

No description available.