Process Modelling as a Help to Decision-Making
for C.E. Implementation
Corresponding Author: Dubois, Anne-Marie
Author(s): Dubois, A.M., and Jägbeck, A.
Organisation(s): CSTB (France)
Concurrent Engineering is well recognised as a source of improvement of productivity and quality in the Construction industry. However, its implementation is progressing slowly. Furthermore, a major question is “how to appropriately apply these technologies within time and budget constraints? What costs and what risks?”

A successful concurrent engineering approach requires a in-depth understanding of the construction process and “modus operandi”. The way information is generated, shared and archived in the « real world », is the most critical aspect of the daily practice. Managers need to know more about their data capture, processing and transmission, as well as the bottlenecks that infringe the efficiency and quality of this process. By the way of adequate modelling methods it is possible to provide them with comprehensive views of the activities and information flows.

The paper reports on several « real projects » modelling works made within the frame of European projects, such as the benchmarking study led by Construct-IT in SCENIC. The work was based on interviews with construction site personnel. A questionnaire and a general diagram of the global construction process were drafted by the research team. It was decided to use the IDEF0 formalism for this first task. An attempt was made to have one common « A0 » diagram (top level process decomposition) so as to result in comparable process descriptions and evaluations. The focus was on the information circuits all the process along

The main lines of the diagrams were presented to the interviewees and then detailed process models were produced using the same formalism, and discussed with different persons on the project sites. The persons (e.g., designers, architects, contractors and site managers) were very interested by the diagrams. IDEF0 diagrams are easy to read and to understand, which is a necessary condition for the success of the work. The different diagrams have given the readers a methodological support to identify the main data flows. Though several extensions were proposed by the research team and the participants, e.g., naming conventions, colours, symbols.

However the process decomposition has generated many discussions. It is difficult for users to accept/recognise a description of their process which sounds too straightforward while their own world is full of crossroads. Also information processes tend to be less fixed than industrial production even if a lot of the components are routine-type. Thus different alternatives may exist at the same time. When going to very detailed diagrams it turns that it would look like a bowl of spaghetti due to the
 numerous complex nested levels.

A provisory conclusion would be that improved process models are necessary, but they must be easy to read and convincing. They must be flexible so as to draw correctly the real word of the construction process. Such models should raise high awareness within the industry on the IT capabilities and the benefits of its appropriate use for CE implementation..