טכניון מכון טכנולוגי לישראל
הטכניון מכון טכנולוגי לישראל - בית הספר ללימודי מוסמכים  
Ph.D Thesis
Ph.D StudentSoffer Pnina
SubjectA Methodology for Adapting an ERP System to the Needs of
an Enterprise
DepartmentDepartment of Industrial Engineering and Management
Supervisors Professor Boaz Golany
Professor Dov Dori


Abstract

Enterprise Resource Planning (ERP) systems are off-the-shelf software packages that support most of the key functions of an enterprise.

One of the main problems in ERP implementation projects is how to align the software package with the business processes of the enterprise implementing it.  The research proposes a requirement-driven approach to the alignment problem, which matches a model of the defined enterprise requirements with a model of the ERP system capabilities. Both models address business issues and apply the Object-Process modeling approach (OPM).

The requirements defined for such off-the shelf information systems, rather than being a complete specification, should allow flexibility in matching with the existing ERP package. The research identifies the essential elements of the requirements as business processes, rules, objects and required system services. Formal definitions of these elements are provided, based on the Bunge-Wand-Weber (BWW) ontology, and form a framework for evaluating requirement modeling tools. The evaluation framework is applied on OPM, whose expressive power is found adequate for specifying the requirements.

The ERP system model represents the entire scope of the ERP system’s functionality and the alternative business processes it supports, as well as the interdependencies among them.

The research presents a generic reverse engineering process for developing this model, and analyzes the desired properties of a suitable modeling language. Using these criteria, OPM is evaluated and applied for generating a detailed ERP system model.

Aligning the ERP system and the enterprise requirements is suggested as an iterative process, which employs an algorithm that matches the two models. The algorithm is able to identify possible matches between the two models and evaluate the gaps between them despite differences in their completeness and detail level. It considers interdependencies among the ERP options, and provides the enterprise with a set of feasible combinations of requirements that can be satisfied by the ERP system.

The feasibility of the suggested alignment approach is evaluated by applying it in an experimental study. The study simulates the alignment process using requirements defined in a real-life ERP implementation project and validates the results by comparing them with the real decisions made. The results of the study establish the viability of the approach for providing a satisfactory solution to the problem of aligning an ERP software package with an enterprise business model.