Digital library of construction informatics
and information technology in civil engineering and construction


Paper w78-2000-497:
A intelligent knowledge-based system for capturing projects’ performance and initiating tendering strategies

Facilitated by the SciX project

Kaka A P

A intelligent knowledge-based system for capturing projects’ performance and initiating tendering strategies

Abstract: "The paper will describe a knowledge-based system developed to capture, process and analyse records of past performance in ordered to help contractors to initiate future strategies and actions. A detailed study of the of a major contractor’s Management Information Systems showed, inter alia, that the “Cost Value Reconciliation” forms (CVR’s), produced on a monthly basis by the contractor’s Surveyor, held a large amount of data unused by the contractors except for the purpose of performance control. Examination of the data revealed that if processed and stored in a central database, it could provide invaluable source of information to contractors for analysing performance and initiating strategies both on the project level and the company level. A subsequent survey of fifteen other contractors revealed that substantially identical CVR procedures were in universal use. It was therefore decided, in order to facilitate the adoption of the proposed system by contractors, to use a CVR format as a data capture facility for the system. Actual CVR sheets used by different contractors were studied, The type of records used in these sheets were examined in terms of their usefulness to management in terms of measuring progress and initiating future actions other than cost control. Further variables that are not currently in use in CVR sheets were introduced. A mathematical model was developed to process these monthly records into useful information subsequently called performance variables (i.e. variables used to measure the contractor’s performance with respect to the project). Finally, further variables were introduces to the system in order to facilitate the sharing of information between different projects. These variables were called Contract Classification Variables (i.e. variables used to describe the project) and contract performance variables) and included nine criteria by which a contract is defined or grouped (e.g. method of tendering, method of procurement etc.). These criteria were identified by contractors as the most important factors influencing contracts’ characteristics and performance. The Contract Performance Variables are the information that can be extracted from the CVR sheets and used to form new strategies (e.g. rate of mark-up on different projects, payment delays etc.). The model was developed in such a way that when a contract is started, the contractor enters the classification and the performance variables in the Individual Project Module. As the contract progresses and actual data become available, the contractor starts to fill the CVR sheets on a monthly basis. When the contract is completed, a the model process the CVR sheets and as a result summarise the performance of the project in terms of the performance variables and the data (including contract classification variables) are sent to the central Database. When a new contract is considered, the contractor defines the project in terms of its classification variables. The model queries the Database for the characteristics of past projects that match the same classification. Once the data is retrieved and processed a set of contract performance variables is predicted for that particular contract. The above method will work as long as adequate similar past projects are found and retrieved from the Database. However, finding adequate data is not always possible, particularly in the early years of applying the proposed model. Also, certain classification variables are not finite in terms of the options available (such as the client for the contract). An Intelligent Data Retrieval system has therefore been developed to overcome this problem. This paper will also explain this system and how the knowledge behind it was elicited."



Full text: content.pdf (301,794 bytes) (available to registered users only)

Series: w78:2000 (browse)
Cluster: papers of the same cluster (result of machine made clusters)
Class: class.retrieve (0.022609) class.processing (0.020626) class.analysis (0.019940)
Similar papers:
Sound: read aloud.

Permission to reproduce these documents have been graciously provided by Icelandic Building Research Institute. The assistance of the editor, Mr. Gudni Gudnason, is gratefully appreciated


hosted by University of Ljubljana



© itc.scix.net 2003
Home page of this database login Powered by SciX Open Publishing Services 1.002 February 16, 2003