Integrated SEM-FTOPSIS framework for modeling and prioritization of risk sources in medical device development process

Abstract

PurposeThe purpose of this paper is to outline and prioritizes risk sources in medical device development (MDD) process using an integrated “structural equation modeling” (SEM) and fuzzy “technique for order performance by similarity to ideal solution (TOPSIS)” framework.Design/methodology/approachRisk sources which deter MDD process are explored through literature review. Initial structural model is proposed, factor loadings are determined by exploratory factor analysis and model fit is established by confirmatory factor analysis. Further, the sources are ranked using FTOPSIS, and sensitivity analysis is carried to check robustness of results.FindingsThe sources of risks have catastrophic effect on MDD process. The initial SEM model developed based on survey of experts is found reliable and valid which breaks up the risk sources into three categories – internal sources of risks, user-related sources of risks and third-party-related sources of risks. The risk sources are ranked and prioritized based on perspective of experts from the categories using FTOPSIS; unmet user needs/requirements is found as the most important source of risk. Results of sensitivity analysis confirm that the factors are relatively less sensitive to criteria weights confirming reliability of initial solution.Research limitations/implicationsThe proposed methodology combines qualitative and quantative approaches, making it little complex and lengthy, but results in dual confirmation.Practical implicationsThe outcomes of this research will be of prime use for MDD industries to mitigate risk sources. It will help to increase the success rate of MDD.Originality/valueIntegrated SEM-FTOPSIS provides a unique and effective structural modeling-based decision support tool. The framework can be effectively utilized in other domains, and failure events of medical devices can be potentially controlled by applying risk mitigation measures.