Dynamic Object Model and Adaptive Workflow

One of the problems with current workflow systems is their limited support for dynamic environments and evolving product and process models. A 1997 study of trends in workflow management concluded that exception handling (i.e., dynamically modifiable processes) and object-oriented views of workflow definitions deserve “serious attention” from researchers [Moh98]. But why are dynamic models so important for adaptive workflow? In [HC93], Hammer and Champy identify three factors that characterize modern businesses: (i) customers take charge, (ii) intense competition, and (iii) constant change. In workflow terms, customers in charge means support for ad-hoc processes. Constant change translates into dynamic process models. In a world of intense competition where “nothing is permanent except change itself” [Cha96], support for ad-hoc processes and dynamic process models are requirements for a succcessful business. Scientists and engineers who work in computerized environments use “scientific” workflow systems. They also require flexible models for process definition and execution [BBG98]. The outcomes of scientific processes evolve as the experiment unfolds. It is difficult to determine in advance the structure of the entire process. Scientific workflow systems must allow workflows and product data to evolve at runtime.