Level: Advanced
TCM Section(s)
7.3. Cost Estimating and Budgeting
10.4. Project Historical Database Management
Venue: 2024 AACE International Conference & Expo
Abstract: Project costs consist of both direct and indirect costs. Indirect costs are those not directly assignable to the end-product or its progress. These include overhead costs for the support teams, general labor costs, transportation costs, insurance, taxes, etc. Indirect costs are often more challenging to control or determine because they represent general cost of the organization generated in favor of more than one project. Estimating these costs can be complex due to variations based on project scope, size, location, and so on.
In this work, a substantial amount of cost data from actual legacy projects analyzed was collected at the work breakdown structure (WBS) level and statistically analyzed. Consequently, recommendations for estimating indirect costs at the line-item level were established based on the data and statistical model.
In the statistical model, scaling factors were introduced to estimate indirect costs for projects across eight different cost bands. Applying the “average - before - fit” methodology, which statistically normalized the fluctuations and random data distribution caused by the heavier weight of small projects, resulted in a significant improvement in the goodness-of-fit, with the R-squared value increasing from approximately 0.1 to around 0.9.
The resulting indirect cost percentage recommendations have been integrated into the estimating tool and have been tested with actual project costs benchmarks. The variance between the calculated indirect costs, based on the newly determined percentages, and their actual benchmark costs was found to be less than 5%.
TCM Section(s)
7.3. Cost Estimating and Budgeting
10.4. Project Historical Database Management
Venue: 2024 AACE International Conference & Expo
Abstract: Project costs consist of both direct and indirect costs. Indirect costs are those not directly assignable to the end-product or its progress. These include overhead costs for the support teams, general labor costs, transportation costs, insurance, taxes, etc. Indirect costs are often more challenging to control or determine because they represent general cost of the organization generated in favor of more than one project. Estimating these costs can be complex due to variations based on project scope, size, location, and so on.
In this work, a substantial amount of cost data from actual legacy projects analyzed was collected at the work breakdown structure (WBS) level and statistically analyzed. Consequently, recommendations for estimating indirect costs at the line-item level were established based on the data and statistical model.
In the statistical model, scaling factors were introduced to estimate indirect costs for projects across eight different cost bands. Applying the “average - before - fit” methodology, which statistically normalized the fluctuations and random data distribution caused by the heavier weight of small projects, resulted in a significant improvement in the goodness-of-fit, with the R-squared value increasing from approximately 0.1 to around 0.9.
The resulting indirect cost percentage recommendations have been integrated into the estimating tool and have been tested with actual project costs benchmarks. The variance between the calculated indirect costs, based on the newly determined percentages, and their actual benchmark costs was found to be less than 5%.