Reliability analysis of marine pilots using advanced decision making methods

Abstract

Seaports play a significant role in global logistics networks, contributing to the efficiency of both national and international economic growth. Dramatic changes in the supply chain encourage ports to maintain effective integration when delivering services. Ports are thus parts of complex systems operating in uncertain operational environments. Accident investigation shows that there has been a significant increase in marine accidents contributed to by human error during marine pilotage operations. The human element has been identified as a major critical factor for most operational failures. Therefore, an adequate understanding of the key factors influencing pilot reliability plays a vital role in all high-risk industries, among which maritime operations are included. This study aims to develop a new quantitative marine pilot reliability assessment methodology, known as the Marine Pilot’s Reliability Index (MPRI). The MPRI seeks to help decision makers in identifying the effects of certain factors on pilot reliability. Although human reliability has been investigated in different disciplines, there is no consensus on the selected criteria. Therefore, in this study, the researcher employed a hybrid research approach, comprised of qualitative and quantitative approaches in a sequential exploratory approach to elicit the key factors that are considered dominant in maintaining the reliability of a marine port pilot. This was conducted through a series of investigation tools such as field observation, semi-structured focus-group interviews, and port pilotage accident data analysis. This step culminated in a composite of four main criteria with thirteen sub-factors, which pilots considered dominant to their reliability. These factors are arranged in a hierarchal order forming the new developed MPRI. To ensure the applicability of the identified MPRI factors, the researcher applied a Delphi technique in examining the degree of agreement among experts towards the identified MPRIs. Two rounds of questionnaires were conducted. The results obtained show a high degree of agreement among experts towards the identified factors. This is followed by the application of the analytical hierarchal process (AHP) approach to determine the relative weights of all identified criteria. The second approach, a new conceptual MPRI interdependency model is constructed using a hybrid approach of a fuzzy decision-making trial and evaluation laboratory (FDEMATEL) and an analytical network process (ANP). This hybrid approach helps to deal with inherent uncertainties and highlights the degree of interdependences in the developed MPRIs. To examine the feasibility of the proposed model and determine the outputs from this research, the researcher employed a fuzzy evidential reasoning (FER) for solving multiple criteria decision-making (MCDM) problems in conjunction with the aforementioned approaches to empirically assess the reliability of a marine pilot. The application of FER helps manage uncertainties resulting from the nature of operations. Three senior marine pilots have been assessed using the developed reliability assessment tool. The results reveal the novelty of this assessment tool in offering an effective and flexible reliability assessment and a diagnostic instrument for decision makers to predict a reduction in a pilot’s reliability. The developed model is partially validated using a sensitivity analysis. The novelty of this work offers a foundation towards assessing the reliability of marine pilotage operations using risk-based methodologies with variance techniques to facilitate the acquisition of qualitative and quantitative data and to ensure safe and efficient port operations.