ABSTRACT
Traditional mathematical methods such as differential equations formalism have been used for centuries as the main tools for the analysis of complex systems in varied areas. As opposed to the general solutions provided by analytical methods, simulation-based techniques are effective methods for designing and analyzing complex systems while providing individual solutions for particular problems. Computer simulation is a technique of representing the real world by a computer program to manipulate that representation in such a way that it operates on time or space. A model is a simplified representation of the real world at some particular point in time or space intended to promote the understanding of its behavioral properties. Therefore, Modeling and Simulation (M&S) is the use of models, including emulators, prototypes, simulators, and either statically or over time, to develop data as a basis for making managerial or technical decisions in the design and analysis of complex systems. Such complex systems represent part of real-world or human-made systems like manufacturing systems, management systems, transportation systems, urban traffic light systems, education systems, military systems, hardware systems, and social systems. One of the areas in which M&S has gained tremendous popularity in these last decades is the domain of healthcare. Considerable efforts have been made about simulation-based study of healthcare systems (HS) as witnessed by a huge amount of work published in recent years. Simulation in HSs has a broad application around different disciplines such as clinical simulation, operational simulation, managerial simulation, and educational simulation. However, the domain of healthcare is characterized by a high degree of complexity and a diversity of perspectives, and modelers are often confronted with the challenge of formulating a simulation model that captures this complexity in a systematic and manageable manner. Frequently used methods to address the problems include discrete event simulation, optimization techniques, goal programming, and data envelopment analysis. We argue that none of the existing works provides an attempt to holistic view modeling of healthcare systems. Instead, the diverse perspectives of healthcare systems are studied in isolation and focus on specific aspects like allocation of resources, disease outbreak control, and population dynamics using specific formalisms. Furthermore, unit-specific simulations are predominant in the literature. Such units include outpatient clinics, A&E (Accident and Emergency Departments), and inpatient facilities. As it turns out, answering questions concerning behavioral properties of the overall system becomes difficult and therefore not sufficient for an efficient design and analysis of the system under study. To address the identified issues, this thesis suggests investigating the domain of HS using M&S with a whole set of activities: (1) an Ontology-based modeling framework, real-world semantics, and a formal specification of the core concepts and their relationships in healthcare simulation, for capturing modeling knowledge in a reusable and interoperable manner, (2) a Multi-perspective M&S Framework for HS, (3) an approach for holistic analysis of HS based on the multi-perspective framework and systematic integration of simulation processes to form an integrated whole, and (4) a formalization of the integration approach through a concept based upon Discrete Event Systems Specification (DEVS) formalism called Parameterized DEVS, whereby concurrent simulation processes cause live update through output-to-parameter integration. vi A top ontology for M&S is developed based on an extensive literature review and provides a framework that covers all the definitions of important concepts in the domain of healthcare systems and the descriptions of their properties. An ontology is a formal specification of a conceptualization that can be used to rigorously define a domain of discourse in terms of classes/concepts, properties/relationships, and instances/individuals. The objective of the Ontology is to document knowledge that is formalized -machines and computers understand-, and assists in communication between humans, achieves interoperability, and facilitates communication among software systems. A multi-perspective approach to modeling and simulation of HSs allows for defining different perspectives that are integrated to form a holistic. The integration between the isolated perspectives is done through concurrent simulations. Hence, studying HSs through multi-perspective modeling provides multiple levels of explanation for the same system. A methodology for a “loose” integration enables independent simulation processes of disparate concerns in HS to exchange live updates of their mutual influences. Such an approach takes the results obtained closer to the reality of the interactions between health phenomena and helps stakeholders gain a holistic understanding of the whole healthcare system while deriving more realistic decisions. A multi-formalism modeling approach to effectively capture the different concerns encapsulated into the multi-perspective framework is built to accommodate the diverse familiarities of the experts with modeling formalisms, reuse of existing models, and easiness to reproducing realities using specific formalism for specific perspectives accordingly. To realize that, we place the multi-formalism modeling feature at the top layer of the proposed framework and support it with the Model Driven Engineering (MDE) approach through which a co-simulation or a formalism transformation can be carried out As a standards-based approach to system development, the concept of OMG’s (Object Management Group) Model Driven Engineering (MDE) is a de facto theoretical unit guiding the investigation and definition of the level of abstractions of a given system of interest. It aims at the development of source models of HSs and transforming them to multiple levels of abstraction until we get to a code level while at the same time increasing the power of models. Finally, a case study is used to illustrate the application of our framework where models are built in isolation according to each perspective and their simulation results are integrated to form a complete whole. The outbreak of Ebola in Nigeria in 2014 is used as a running example.
A MODEL-DRIVEN ENGINEERING FRAMEWORK FOR HEALTHCARE SYSTEMS MANAGEMENT, GETTING MORE PROJECT TOPICS AND MATERIALS