Loading...

Enabling Flexibility in Workflow Systems at Runtime

Tahmooresi, Hamed | 2014

605 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 46438 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Mirian Hoddeinabadi, Hassan
  7. Abstract:
  8. Nowadays, small and big businesses do their best to automate their business process by using a workflow management system. Therefore, these systems play a key role in IT. By expansion of the business and ever-changing rules and procedures of organization, the need for consecutive changes to workflow management systems arises. The most challenging part of this change is the state of running instances. Previous solutions put the challenges of migrating instances, dealing with their inconsistencies and runtime errors on the users. It is worth noting that there are a few tools which consider instance migrations, but most of them lack a comprehensive viewpoint. Related solutions are often based on old languages which are not practical for new languages. The goal of this research is to provide a new solution to solve the flexibility problem at runtime in workflow management systems. To this purpose, we develop a high-level process by using a goal-driven approach. Each phase in this process needs some products as its input and generates some outputs. In this solution, all algorithms and low-level tasks are replaceable. In fact, each phase is composed of multiple method chunks which can be replaced by other chunks from a repository. To increase clarity and tangibility of our process, a high-level example independent of modeling language is presented for each phase. Users can utilize these examples or replace them with their own solutions. This approach results in a dynamic and extensible solution which can be customized based on the modeling language, company situation and the process sensitivity
  9. Keywords:
  10. Workflow ; Flexibility ; Business Process Management System ; Running Instances ; Adaptation Pattern

 Digital Object List

 Bookmark

No TOC