Fuzzy Agent-based Modeling of Competency and Performance Measures of Construction Organizations

• Author / Creator

  Gebretekle, Yisshak T.

• Construction competency measures refer to a combination of the skills, knowledge, technologies, processes, and practices that a construction organization uses to improve its performance and the performance of its projects. Performance measures in construction are used to assess the effectiveness, competitiveness, and profitability of organizations based on ongoing or completed actions or processes. Previous studies have identified and developed separate measures for construction competency and performance at both the project and organization levels. However, in practice, construction organizations are project-based, and they develop competencies and performance through project execution. Project-level and organizational competencies are viewed as having a two-way relationship, where organizational competencies drive, orient, and support multiple projects and are simultaneously constantly investigated and redefined through practices brought by project-level competencies. Thus, project-level and organizational competencies appear inseparable in construction organizations. However, assessing the impact of a multilevel competency on performance is challenging because of the complex interaction between competencies and performance measures and the subjective uncertainty that exists in their measurement. Hence, an integrated framework is needed that provides a well-defined, structured model of categories of competencies that coexist in construction organizations and their projects, and the impact of these competencies on performance measures. This multilevel modeling framework should allow both competencies and performance to develop through project execution, meaning it needs to be a bottom-up modeling framework. Such a multilevel modeling framework is not found in the literature. Therefore, a significant gap exists for researchers seeking to improve competency and performance modeling in construction organizations.
  To bridge this gap, this thesis aims to develop a hybrid fuzzy logic and agent-based model (FABM) that can handle the complex, dynamic, and subjectively uncertain nature of construction competency and performance measures to analyze construction competencies as inputs and predict multiple performance measures simultaneously. First, a comprehensive set of project level and organizational competency and performance measures were summarized and updated from the existing literature. Then, based on the proposed multilevel competency and performance framework, a FABM model was developed that constitutes the multi-agent environment corresponding to the hierarchical category of competency measures established. The fuzzy c-means (FCM) clustering and fuzzy inference system (FIS) were used to develop fuzzy membership functions and the initial behavioral and interaction rules of the competency agents based on the available data of project level and organizational competency measures. The parameters of the FCM clustering were further optimized, and a final set of FISs were established for the behavioral and interaction rules of the competency agents. FIS receives input competency measure variables from the multi-agent environment and delivers the predicted behavior of each agent as well as emergent performance measures. Existing construction competency and performance datasets were used to validate and verify the FABM model, with encouraging results in multilevel competency-based performance prediction in construction organizations. The contribution of this study is providing a systematic bottom-up modeling approach to measure and assess competencies at the project and organization levels and mapping these multilevel competencies to construction performance measures. Furthermore, the outcomes of this study are expected to support construction practitioners by providing a set of comprehensive hierarchical competency and performance metrics at the project and organization levels for managerial actions taken to identify, construct, and develop competency models to assess performance at both levels.

• Subjects / Keywords

  • Fuzzy logic
  • agent-based modeling
  • competency measures
  • construction
  • performance measures
  • fuzzy inference system

• Graduation date

  Fall 2023

• Type of Item

  Thesis

• Degree

  Master of Science

• DOI

  https://doi.org/10.7939/r3-g9nh-jn16

• License

  This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.