Abstracts – Browse Results
Click on the titles below to expand the information about each abstract.
Viewing 1 results ...
Kishk, M, Pollock, R and Okhuahesogie, E (2004) Sensitivity of simulated Whole-Life Costs to input probability distributions. In: Khosrowshahi, F (Ed.), Proceedings 20th Annual ARCOM Conference, 1-3 September 2004, Edinburgh, UK. Association of Researchers in Construction Management, Vol. 1, 115–24.
- Type: Conference Proceedings
- Keywords: Decision-making; Monte Carlo simulation; probability distributions; sensitivity analysis; whole-life costing
- ISBN/ISSN: 0 9534161 9 4
- URL: http://www.arcom.ac.uk/-docs/proceedings/ar2004-0115-0124_Kishk_Pollock_and_Okhuahesogie.pdf
- Abstract:
In a typical Monte Carlo simulation (MCS), it is required to assign probability distribution functions (PDFs) for uncertain parameters. Due to lack of or irrelevance of historical whole-life costing (WLC) data, these functions are usually assumed. The research work that underpins this paper aimed to identify the significance of errors in various input parameters in a WLC exercise and to identify critical parameters in assigning a probability distribution. In carrying out the study, a number of case studies have been identified and analysed using a risk assessment software. Besides, a number of metrics to assess the impact have been identified and employed. Typical results show that varying the type and range of input PDFs only affect the certainty level of simulated WLC results while variations in the standard deviation affected the risk ranking of uncertain input variables. The most significant impact was that due to variations in the mean of input PDFs, which does not only affect the certainty of the decision, but can even change the decision altogether.