- 155 views
- 181 downloads
Hybrid Neuro-fuzzy Model for Construction Organizational Competencies and Performance
- Author / Creator
- Getaneh Gezahegne Tiruneh
The construction industry is dynamic and complex that demands continuous quality, productivity, and performance improvement; making it challenging to achieve organizational success, superior performance, and competitive advantage. Organizational competencies have a significant influence on performance; hence, it is vital that construction organizations assess and enhance their competencies in order to improve performance. In addition, relating organizational competencies to performance is essential to identify target areas leading to improved performance. Furthermore, the variables that characterize organizational competencies and performance are both quantitative and qualitative in nature, and thus require measurement methods and modeling techniques such as artificial intelligence (AI) that can handle both variable types. However, stand-alone AI techniques have limitations for handling complex real-world problems. For instance, fuzzy systems are strong in reasoning and inference and explicit knowledge representation while weak in learning capabilities. On the other hand, artificial neural networks (ANNs) have powerful learning ability while poor in reasoning and inference. Thus, hybrid modeling approaches that combine two or more AI methods such as neuro-fuzzy systems (NFS), that combine the learning power of ANNs and functionality of fuzzy systems (i.e., improving reasoning and inference and explicit knowledge representation), are viable options used for modeling and solving practical real-world problems such as predicting performance.
NFS models have proven to be very effective for a wide range of real-world applications in construction owing to their robust, fast, and effective characteristics for solving complex problems. However, the application of different types of NFS models have some limitations such as (1) handling multiple outputs that are common in real-world construction processes and practices, and (2) suffering from local minima and poor generalization that may lead to provide less accurate results and/or inadequate explanations for problems. Therefore, a hybrid NFS that combines evolutionary optimization technique i.e., genetic algorithm (GA) and multi-output adaptive neuro-fuzzy inference systems (MANFIS) is developed in this research to analyze multiple inputs and multi-outputs, that relate organizational competencies to performance, and predict multiple organizational performance metrics.
A systematic review and detailed content analysis of selected articles was conducted to identify, categorize, and rank organizational competencies affecting organizational performance. The categorization of competency and performance metrics, verified by the focus group, provides organizations with a systematic method to evaluate their competencies and improve their performance. The list of organizational competencies and performance metrics were piloted tested with a construction company prior to the data collection to ensure construct validity and the reliability of evaluation and measurement techniques used for data collection.
This research provides both researchers and construction industry practitioners a hybrid NFS modeling approach to analyze multiple organizational competencies as model inputs, relate them to performance, and predicting organizational performance. The hybrid NFS model enables to identify potential competencies for performance improvement, which provide organizations as well as construction practitioners with insight into targeted areas for future investment and expansion strategies in order to improve organizational performance, which further helps them to make the best decisions. Additionally, the hybrid NFS model has a great advantage since it can predict multiple organizational performance metrics simultaneously rather than developing independent models for each output.
This thesis is an original work by Getaneh Gezahegne Tiruneh. The research project, on which this dissertation is based on, received research ethics approval from the University of Alberta Research Ethics Board, Project Name “Fuzzy Hybrid Techniques for Competency Modeling for Construction Organizations and Projects”, Study ID: Pro00068907, approved on November 04, 2016. This research was funded by the Natural Sciences and Engineering Research Council of Canada Industrial Research Chair in Strategic Construction Modeling and Delivery (NSERC IRCPJ 428226–15), which is held by Dr. Aminah Robinson Fayek.
Parts of Chapter 2 of this thesis have been published in Automation in Construction: Tiruneh, G. G., A. R. Fayek, and S. Vuppuluri. 2020. “Neuro-fuzzy systems in construction engineering and management research.” Autom. Constr., 119: 103348. https://doi.org/10.1016/j.autcon.2020.103348. Chapter 3 and parts of Chapter 2 of this thesis has been accepted for publication on May 26, 2020, and Published on the web on May 29, 2020, in the Canadian Journal of Civil Engineering: Tiruneh, G. G. and A. R. Fayek. 2020. “Competency and performance measures for organizations in the construction industry.” Can. J. Civ. Eng., 50 manuscript pages, https://doi.org/10.1139/cjce-2019-0769. Chapters 5 and Chapter 6 and parts of Chapter 2 of this thesis have been submitted for publication in Journal of Computing in Civil Engineering: Tiruneh, G. G. and A. R. Fayek. 2021. Hybrid GA-MANFIS model for organizational competencies and performance in construction. J. Comput. Civ. Eng., 43 manuscript pages, submitted Jan. 15, 2021. I was responsible for the data collection and analysis, as well as the composition of the three manuscripts. Dr. Aminah Robinson Fayek was the supervisory author and was involved with concept formation and composition of each of the three manuscripts.
- Graduation date
- Spring 2021
- Type of Item
- Doctor of Philosophy
- 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.