The use of intelligent simulation in cost-time forecasts for housing rehabilitation works
The research aimed to investigate the feasibility of using intelligent simulation for construction cost-time forecasting in order to provide an aid to strategic decision-making in housing rehabilitation work. Essential features were anticipated as being the provision of probabilistic forecasts and the ability to reflect the operational consequences of alternative design and management strategies.
The intelligent simulation envisaged was essentially a hybrid of stochastic simulation and knowledge based systems such that:
• an operational plan for the project would be generated automatically by the system, using knowledge drawn from expert planners;
• stochastic simulation would be used to evaluate project time and cost from the production plan, thus providing probabilistic forecasts;
• site management control of the production process would be emulated by interaction of the simulation with a knowledge base representing the expertise of the site managers;
• a graphics-based user interface would permit real time monitoring and interaction with the simulation.
Low-rise multiple-unit housing rehabilitation projects of a clearly defined type were used as the vehicle for this investigation, as these projects are characterised by high levels of uncertainty and sensitivity of cost and time to operational factors.
The approach adopted was to seek to build a trial system or concept demonstrator. As the aim was to investigate feasibility, no attempt was made to acquire full domain knowledge, effort being concentrated upon establishing the structure and nature of the knowledge to be encapsulated, and determining appropriate methods of knowledge representation and system architecture. Conceptual design of the system was completed within the grant period, but some work remains to be done to achieve an operational concept demonstrator.
The main findings were as follows:
1. A system integrating knowledge-based approaches with stochastic simulation is readily achievable within an object-oriented structure.
2. Automatic operational planning for repetitive works of these kind is also readily achievable, within an object oriented structure.
3. The domain of planning expertise examined proved suitable for KBS development. The techniques and heuristics employed are well established and consensus between experts is strong.
4. The domain of site management expertise presented problems; this knowledge appears difficult to both acquire and represent. Only limited progress was made with the simple and direct knowledge elicitation techniques used: there is some evidence that more sophisticated indirect techniques may perform better but it was not possible to test this within the grant period.
5. The experts in housing rehabilitation that were consulted strongly reinforced the view that there is a real need for tools to support both strategic decision making and planning on this type of project.
1. Feasibility of the intelligent simulation approach
6. The research indicated that current levels of hardware and software technology are capable of supporting the development of a working intelligent simulation-based construction cost and time forecasting system, where:
• an object-oriented structure is adopted;
• the construction projects involved are repetitive, multiple-unit schemes of a type for which a well established body of planning expertise exists;
• the variety of construction work to be handled by the system is limited;
• a very limited and crude representation of site management control decisions and actions is acceptable.
Citation countsare sourced monthly fromand citation databases.
These databases contain citations from different subsets of available publications and different time periods and thus the citation count from each is usually different. Some works are not in either database and no count is displayed. Scopus includes citations from articles published in 1996 onwards, and Web of Science generally from 1980 onwards.
Citations counts from theindexing service can be viewed at the linked Google Scholar™ search.
Full-text downloadsdisplays the total number of times this work’s files (e.g., a PDF) have been downloaded from QUT ePrints as well as the number of downloads in the previous 365 days. The count includes downloads for all files if a work has more than one.
|Subjects:||Australian and New Zealand Standard Research Classification > BUILT ENVIRONMENT AND DESIGN (120000) > BUILDING (120200) > Quantity Surveying (120203)|
|Divisions:||Past > QUT Faculties & Divisions > Faculty of Built Environment and Engineering|
|Copyright Owner:||Copyright 1991 (The authors)|
|Deposited On:||13 Sep 2007|
|Last Modified:||09 Jun 2010 22:45|
Repository Staff Only: item control page