WP023: The Haztimator: Knowledge-Based Time and Cost Estimating of Hazardous Waste Remediation Projects

This paper presents a new approach to conceptual cost estimating that will enable better cost control starting from the early stages of the design process. Our approach is based on the use of model-based reasoning tools to generate plausible alternative designs automatically at conceptual and preliminary stages of the design, with the limited amount of data available at those stages. We believe that these plausible detailed designs can be accurately estimated and then refined during later stages of design serving as "defending champions" against which every succeeding detailed design - or "challenger" may be compared. Further, the costs associated with these designs can be stored with semantic links to the design components in the design model, and can be refined as the design evolves. We claim that such an approach will have substantial impacts on the way that civil engineers can estimate future projects and control their costs.

This paper introduces a model-based design and cost estimating system, "Haztimator", which was implemented to apply this approach to the hazardous waste engineering domain. We will first describe the problems associated with the traditional approach to conceptual cost estimating focusing on the unique challenges of the Hazardous Waste domain. A section on Haztimator and its implementation environment will follow this introduction. Then, we will describe the potential implications of our approach on the design and project management process, and will provide some predictions about the impacts of design automation on the organizational structure of civil engineering firms. The final section of the paper will discuss possible future extensions to our proof-of-concept prototype.