A group AHP-TOPSIS framework for human spaceflight mission planning at NASA

Abstract

Human spaceflight mission planning is a complex task with many interacting systems and mission phases. Analog missions are Earth-based science missions whose purpose is to help understand the complexities inherent in future human spaceflight missions. The goal of performing an analog mission is to prepare crewmembers and support teams for future space missions in a low risk-low cost environment by repeatedly testing vehicles, habitats, and surface terrain simulators. This study presents a group multi-attribute decision making (MADM) framework developed at the Johnson Space Center (JSC) for the Integrated human exploration mission simulation facility (INTEGRITY) project to assess the priority of human spaceflight mission simulators. The proposed framework integrates subjective judgments derived from the analytic hierarchy process (AHP) with the entropy information and the technique for order preference by similarity to the ideal solution (TOPSIS) into a series of preference models for the human exploration of Mars. Three different variations of TOPSIS including conventional, adjusted and modified TOPSIS methods are considered in the proposed framework.

Introduction

The primary goal in multi-criteria decision making (MCDM) is to provide a set of attribute aggregation methodologies that enable the development of models considering the decision makers’ (DMs’) preferential system and judgment policy (Doumpos & Zopounidis, 2002). Achieving this goal requires the implementation of complex procedures. While intuition and simple rules are still favorite decision making methods, they may be dangerously inaccurate for complex decision problems.

Roy (1990) argues that solving MCDM problems is not searching for an optimal solution, but rather helping DMs master the complex judgments and data involved in their problems and advance towards an acceptable solution. Multi-attributes analysis is not an off-the-shelf recipe that can be applied to every problem and situation. The development of MCDM models has often been dictated by real-life problems. Therefore, it is not surprising that methods have appeared in a rather diffuse way, without any clear general methodology or basic theory (Vincke, 1992). The selection of a MCDM framework or method should be done carefully according to the nature of the problem, types of choices, measurement scales, dependency among the attributes, type of uncertainty, expectations of the DMs, and quantity and quality of the available data and judgments (Vincke, 1992). Finding the “best” MCDM framework is an elusive goal that may never be reached (Triantaphyllou, 2000).

The MCDM methods are frequently used to solve real-world problems with multiple, conflicting, and incommensurate attributes. Several authors have used MCDM to effectively solve complex space exploration problems at the National Aeronautic and Space Administration (NASA). Tavana (2003) developed a MCDM model with crisp data to evaluate and prioritize advanced-technology projects at the Kennedy Space Center (KSC). Tavana and Sodenkamp (2009) extended this model with fuzzy data and proposed a fuzzy MCDM model for technology assessment at KSC. Tavana (2004) proposed a MCDM model to evaluate a set of alternative mission architectures for the human exploration of Mars. Tavana et al., 2007, Tavana, 2008 developed two group multi-criteria decision support systems at JSC, a workforce planning system, and an environmental benchmarking system, respectively.

MCDM problems are generally categorized as continuous or discrete, depending on the domain of alternatives. Hwang and Yoon (1981) have classified the MCDM methods into two general categories: multi-objective decision making (MODM) and multi-attribute decision making (MADM). MODM has been widely studied by means of mathematical programming methods with well-formulated theoretical frameworks. MODM methods have decision variable values that are determined in a continuous or integer domain with either an infinitive or a large number of alternative choices, the best of which should satisfy the DMs constraints and preference priorities (Hwang and Masud, 1979, Ehrgott and Wiecek, 2005). MADM methods, on the other hand, have been used to solve problems with discrete decision spaces and a predetermined or a limited number of alternative choices. Churchman, Ackoff, and Arnoff (1954) initially proposed a simple additive weighting method for selecting a business investment policy. The MADM solution process requires inter and intra-attribute comparisons and involves implicit or explicit tradeoffs (Hwang & Yoon, 1981). A detailed analysis of the theoretical foundations of different MCDM methods and their comparative strengths and weaknesses is presented in Larichev and Olson, 2001, Belton and Stewart, 2002, Figueira et al., 2005.

This study presents a group MADM framework based on the analytic hierarchy process (AHP), entropy and the technique for order preference by similarity to the ideal solution (TOPSIS) that were developed for the Integrated Human Exploration Mission Simulation Facility (INTEGRITY) project at the Johnson Space Center (JSC) to assess the priority of a set of human spaceflight mission simulators. The proposed MADM framework integrates subjective judgments derived from the AHP with entropy data and TOPSIS into a series of preference models to prioritize five mission simulators for the human exploration of Mars. The structured framework presented in this study has some obvious attractive features:

• The generic nature of the framework proposed in this study allow for the subjective evaluation of a finite number of decision alternatives on a finite number of performance attributes by a group of DMs.

• The mathematical and computational properties of the models are applicable to a wide range of real-world decision making problems in MADM.

• The information requirements of the proposed framework are stratified into a hierarchy to simplify information input and allow the DMs to focus on a small area of the large problem. This process is also useful for seeking input from multiple DMs.

• Inconsistencies are inevitable when dealing with subjective information from different DMs. The built-in inconsistency checking mechanism of the proposed framework helps to identify inconsistencies in judgments at very early stages of the computation process.

The remainder of the paper is organized as follows. Section 2 presents a brief overview of AHP. In Section 3, we provide a detailed description of three TOPSIS models considered for the proposed framework. Section 4 demonstrates the problem of rank reversal in MADM and TOPSIS through numerical examples. In Section 5, we introduce the INTEGRITY project at NASA. Section 6 presents the details of the group MADM framework proposed in this study along with the results of the INEGRITY problem. Section 7 summarizes our conclusions and future research directions.