Large scale order processing using the navigation plan concept

Abstract

Order processing is an important application in business process management. Two of the fundamental challenges in the design and implementation of order processing systems are: (1) interoperation and integration of heterogeneous and autonomous information systems, and (2) flexible evolution and composition of software components that implement the workflow of order processing. In this research project we propose the Navigation Plan concept as solution of several order processing challenges. The main hypothesis of our research is that the three stages of order processing can be separated, implemented, integrated, and executed through well-defined interfaces that support both the integration of heterogeneous and autonomous information systems and flexible evolution of workflows. The three stages are: first, a client (either human user or another program) generates an order, i.e., a request for information, physical goods, or services; second, the order is validated by going through a series of business steps. This is important since legitimate orders can be fulfilled only when they satisfy the requirements specified in those business steps; third, a validated order is passed along to an appropriate execution process to carry out the request. The representation and execution of business steps in workflow environment have been generated some challenges. One of them is: Which is the best formal foundation and language for controlling workflow patterns? Some of workflow languages advocate the Petri Nets or Process Algebra approaches as formal foundation. However, the implementation of workflow patterns described by Process Algebra remains an open problem. In this research, Process Algebra has been used to represent the semi-formal description of Navigation Plan Definition Language (NPDL). Six research activities are proposed in this research project to deep and strengthen the Navigation Plan concept. (AU)