Analysis of Design-Related Problems from an Industrial Project
Abstract
Historical data from completed construction projects are not always efficiently used when planning a future project because the data are rarely reviewed and synthesized into useful information. In this dissertation, the data from a completed industrial project are examined, with particular emphasis on the design-related problems discovered during the construction phase of the project. The problems were identified and classified into eight categories through the Inter-Office Memos which originated from the construction site. Problems are graphically displayed for each problem category, organization of the problem originator, and discipline for each of the four process areas. The results show that the timing and number of problems for each unit are similar throughout the construction phase of the project. Problems dependencies between problem categories, organization, and discipline are shown through the use of contingency tables. Problem deviations are analyzed by testing the hypothesis that the percentage of problems are linearly proportional to the percentage of discipline progress complete. Equipment and Structural problems followed a pattern which showed problems developing at the end of the discipline's activities. Instrumentation, a changing technology, showed the most scatter, but tended to follow the same type of pattern as Equipment. Forecasting models are developed using regression analysis. Three models are developed to predict the potential number of design-related problems: 1) a model using design manhours, 2) a model using construction manhours, and 3) a model using the discipline's progress schedule. A time series model is tried, but the results indicate that this method is not appropriate to analyze these problems.
Document Details
- Document Type
- Technical Report
- Publication Date
- May 01, 1988
- Accession Number
- ADA196201
Entities
People
- John R. Glavan
Organizations
- Air Force Institute of Technology