An integrated fuzzy MCDM based approach for robot selection considering objective and subjective criteria

Highlights

• A model for the selection of robots by considering both objective and subjective criteria is proposed.

• Justification on selection of criteria is carried out using Fuzzy Delphi Method (FDM).

• Weights of the criteria are found using FAHP.

• Alternatives are ranked by Fuzzy TOPSIS/Fuzzy VIKOR method.

• Ranking identified using Fuzzy VIKOR is closest to ideal solution.

Abstract

Robots with vastly different capabilities and specifications are available for a wide range of applications. Selection of a robot for a specific application has become more complicated due to increase in the complexity, advanced features and facilities that are continuously being incorporated into the robots by different manufacturers. The aim of this paper is to present an integrated approach for the optimal selection of robots by considering both objective and subjective criteria. The approach utilizes Fuzzy Delphi Method (FDM), Fuzzy Analytical Hierarchical Process (FAHP), Fuzzy modified TOPSIS or Fuzzy VIKOR and Brown–Gibson model for robot selection. FDM is used to select the list of important objective and subjective criteria based on the decision makers’ opinion. Fuzzy AHP method is then used to find out the weight of each criterion (both objective and subjective). Fuzzy modified TOPSIS or Fuzzy VIKOR method is then used to rank the alternatives based on objective and subjective factors. The rankings obtained are used to calculate the robot selection index based on Brown–Gibson model. The proposed methodology is illustrated with a case study related to selection of robot for teaching purpose. It is found that the highest ranked alternative based on Fuzzy VIKOR is closest to the ideal solution.

Introduction

Robots are very powerful elements of today's industry and are defined as automatically controlled, reprogrammable, multipurpose manipulators programmable in three or more axes [1]. The recent advancement in automation field has lead to increased usage of robots with distinct capabilities, features and specifications. A company's competitiveness in terms of the productivity of its facilities and quality of its products will be adversely affected by improper selection of robots [2], [3]. The determination of the most appropriate robot considering multiple conflicting qualitative and quantitative criteria has been a difficult task for the decision makers. Literature reveals that multi-criteria decision making (MCDM) methods, production system performance optimization models, computer-assisted models and a general category of solutions are employed in robot selection process [4], [5], [6]. Among these models, MCDM is the most common method for ranking robots and production system performance optimization models are rarely used [5], [6]. Hence, the methodologies for robot selection are divided into three broad categories: (a) MCDM, (b) integrated approaches, and (c) a general category of solutions.

Various MCDM methods reported in the literature for robot selection process are: AHP (Analytic Hierarchy Process), TOPSIS (technique for order preference by similarity to ideal solution), VIKOR (VIsekriterijumsko KOmpromisno Rangiranje), ELECTRE II (ELimination and Et Choice Translating REality), PROMETHEE II (Preference Ranking Organization METHod for Enrichment Evaluation) and DEA (Data Envelopment Analysis). Due to vagueness in the data and ambiguity in decision-making process, fuzzy set theory has been incorporated into MCDM techniques by many researchers for robot selection problem. The selection criteria and the techniques considered by various researchers are presented in Table 1. Among these techniques, TOPSIS and VIKOR methods seemed to be more appropriate for solving the robot selection problem because they have capability to deal with each kind of judgment criteria, having clarity of results and easiness to deal with attributes and decision options [12].

In the integrated approach category, one or more techniques are integrated or combined to select robot for various applications. Table 2 provides the list of works related to integrated approach category along with their selection criteria. The general category of robot selection models includes statistical models, mathematical models and soft computing based models. Table 3 provides the list of works related to general solution category along with their selection criteria. From all the approaches, various criteria considered for selecting a robot for a particular application includes: load capacity, repeatability, horizontal and vertical reach, velocity or speed of travel, degree of freedom, positioning accuracy, maximum tip speed, memory capacity, manipulator reach, velocity ratio, purchase cost, repeatability error, warranty period, vendor's service quality, programming flexibility, man–machine interface, stability, compliance, service contract, drive system, geometrical dexterity, path measuring system, robot size, material of robot, weight of robot, initial operating cost and life-expectancy. The literature review demonstrates that the majority of researchers have considered the selection of proper criteria as important, but they did not justify their selection or analyze their suitability [4]. Hence, a proper methodology to justify the selection of objective and subjective factors has to be incorporated in the decision making process. Fuzzy Delphi Method can be employed to overcome this limitation.

The criteria which are defined in tangible terms are classified as quantitative or objective criteria e.g. cost, load capacity etc. and the criteria that have qualitative definitions i.e., they are unable to be defined in tangible terms are classified as qualitative or subjective criteria e.g. programming flexibility, stability, etc. The classification of criteria into objective and subjective for robot selection is done by only few authors [11], [3], [32], [37]. Even though few researchers have considered this type of classification, weights given to objective and subjective factors are equal. But in real-life applications, the weights for quantitative and qualitative evaluation criteria will be different. Hence in the developed approach, this limitation is taken care off.

Therefore, this paper aims to develop a new decision making method which takes care of suitable objective and subjective criteria selection and proper evaluation of the alternatives treating it as a MCDM problem. The proposed approach integrates Fuzzy Delphi Method, Fuzzy AHP, Fuzzy TOPSIS/Fuzzy VIKOR and Brown–Gibson model. In order to do so, the remainder of this paper is set out as follows. The proposed methodology is described in Section 2. In Section 3, selection of robot for educational purpose is used to illustrate the proposed method. Conclusions are presented in Section 4.