Human Resource Risk Management in Construction Projects Based on the ANFIS Method

Abstract

Human resource risks have a significant impact on the performance of an organization, especially project-based organizations, and the issue of identifying and managing risk is very important in this process. This paper attempts to present an integrated model based on an adaptive neuro-fuzzy inference system (ANFIS) for risk assessment by identifying effective human resource management risks in construction projects. Based on the analyzes performed and determining the likelihood, the severity of the impact, and the rate of occurrence of the risks, a total of 9 risk factors were identified as the most important risks at a high level (unauthorized) and critical importance, including (1) the risk of shortage and lack of maturity and knowledge of specialist staff, (2) job stress, (3) lack of staff motivation, (4) lack of attention to performance management process and feedback, (5) lack of financial resources to allocate rewards, innovative activities, (6) inappropriate payment policies, (7) discriminatory HR practices, (8) lack of technical knowledge and skills, and (9) lack of trust-building among staff. The results of correlation coefficient criterion values to determine the error rate of the two proposed methods showed that random data had a higher correlation coefficient than expert data and the results are satisfactory with acceptable accuracy. Qualitative and quantitative analysis of risks by ANFIS method based on expert opinion and random data showed that if the risks are not present in the projects that are being implemented for the first time or lacking the required number of experts, acceptable results are obtained by adopting and selecting input data based on random data from fuzzy intervals rather than by experts and analyzing them using ANFIS method