Evaluation of an artificial intelligence project in the software industry based on fuzzy analytic hierarchy process and complex adaptive systems

Abstract

PurposeArtificial intelligence (AI) is the most progressive commodity among current information system applications. In-house development and sales of beneficial products are difficult for many software development and service companies (SDSCs). SDSCs have some implicit concerns about implementing AI software development due to the complexity of AI technology; they require an evaluation framework to avoid development failure. To fill the void, this study identified the factors influencing SDSCs when developing AI software development.Design/methodology/approachBased on complex adaptive systems theory, three aspects were developed as the main factors of hierarchy, namely, employees' capabilities, environmental resources and team capabilities. Fuzzy analytic hierarchy process (FAHP) was used to assess the SDSCs' attitude. Based on SDSCs, attitudes toward implementing AI software projects were collected to calculate the hierarchy of factors.FindingsThe outcome of FAHP is used as understanding the key factors of SDSCs for selecting an AI software project, toward the improvement of overall project planning. Employees' stress resistance was considered as a priority for the project, although professional AI skills and resources were also important.Originality/valueThis study suggested three variables developed using complex adaptive systems. This study contributes to a better understanding of the critical aspects of developing AI software projects in SDSCs. The study's findings have practical and academic implications for SDSCs and subsequent academic development, broadening the scope of AI software development research.

Fuzzy Integrated Delphi-ISM-MICMAC Hybrid Multi-Criteria Approach to Optimize the Artificial Intelligence (AI) Factors Influencing Cost Management in Civil Engineering

“…Model development (F12): Developing robust models creates frameworks that accurately reflect project dynamics, facilitating precise cost estimation and resource allocation. Well-constructed models improve project planning, reduce financial risks, and contribute to the successful completion of civil engineering projects within budgetary limits [33,34,42,47].…”

“…Machine learning can learn from historical data to predict resource demands and dynamically adjust allocation strategies. AI-driven systems can maximize productivity, minimize waste, and improve outcomes with limited resources [12,[17][18][19][20][21]44,47].…”

“…AI algorithms can handle data from multiple sources, enabling their seamless integration with existing workflows. This AI-based approach enhances efficiency, interoperability, and data-driven decision-making in civil engineering projects [33,34,47,48].…”

“…• System compatibility (F71): System compatibility ensures the seamless integration of software tools and data platforms, improving project efficiency and reducing operational costs. It facilitates smooth data exchange and communication, fostering collaboration among project stakeholders and ultimately optimizing cost management within budget constraints [33,47].…”

“…The social implications cover broader issues like job displacement, inequality, and autonomy concerns due to AI's adoption. Ethical AI frameworks, guidelines, and governance mechanisms are crucial for managing risks, fostering trust, and promoting ethical practices in civil engineering projects [43,45,47,49].…”