Integration of Rough SWARA and COPRAS in the Performance Evaluation of Third-Party Logistics Providers

The global market growth that is getting faster worldwide has resulted in an increasing logistics function [1]. It causes the supply chain problem to become more complex because it involves transportation and distribution arrangements [2]. Transportation is one of the dominant logistical activities in the supply chain. This activity accounts for a significant total operational cost [3] [4]. The company's logistics system implements several strategies to minimize costs, including using their vehicle, hiring a 3PL provider, and using both options depending on relevant needs [5]. The use of logistics provider services (3PL) in companies is currently a decision that is often used by companies [6] [7] [8]. 3PL provider performance evaluation is a crucial evaluation process in supply chain management [9]. The provider performance evaluation process is used to measure performance and determine follow-up on necessary things [10]. This activity is used to ensure customer needs have been met. ARTICLE INFO ABSTRACT


Introduction
The global market growth that is getting faster worldwide has resulted in an increasing logistics function [1]. It causes the supply chain problem to become more complex because it involves transportation and distribution arrangements [2]. Transportation is one of the dominant logistical activities in the supply chain. This activity accounts for a significant total operational cost [3] [4]. The company's logistics system implements several strategies to minimize costs, including using their vehicle, hiring a 3PL provider, and using both options depending on relevant needs [5]. The use of logistics provider services (3PL) in companies is currently a decision that is often used by companies [6] [7] [8]. 3PL provider performance evaluation is a crucial evaluation process in supply chain management [9]. The provider performance evaluation process is used to measure performance and determine follow-up on necessary things [10]. This activity is used to ensure customer needs have been met.
Currently, the use of logistics service providers in companies has become a decision chosen by several companies. Companies use Third Party Logistics (3PL) to focus more on other essential activities in the company. Evaluating the performance of a 3PL provider is an essential process in determining the performance of a 3PL provider. A wrong evaluation process could lead to company's loss. The main objective of this study was to propose a 3PL performance appraisal procedure. This study integrated the Rough method Step-wise Weight Assessment Ratio Analysis (SWARA) and the Complex Proportional Assessment Method (COPRAS) to assess the performance of 3PL providers. The SWARA Rough method was used to assess the ranking of the criteria. The results of the Rough SWARA ranking were utilized by the COPRAS method to assess supplier performance. A case study was conducted in an animal feed production company in Indonesia. The results showed that there were criteria for product safety; on-time delivery, responsiveness, and flexibility with the greatest weight among the 16 criteria used.
According to Manotas-Duque, et al. [11], 3PL service providers are external companies that manage, control, and support logistics activities. The 3PL provider performance evaluation process needs to determine the criteria needed to suit the company's problem conditions. Several studies were carried out on this problem, as expressed by Yeung [12]. His research evaluated 72 exporters in Hong Kong using four criteria: timeliness of service, price, quality of delivery, and additional services. The same research was also conducted by Mardani and Saptadi [13]. From the two studies above, the performance evaluation of 3PL providers is considered critical in logistics. The 3PL provider performance evaluation begins with identifying criteria by the company's conditions.
Apart from these two studies, several Multi-Criteria Decision Making (MCDM) approaches related to 3PL have been proposed. Some of the proposed procedures include Analytic Network Process, Analytic Hierarchy Process AHP [14] [15], Fuzzy AHP [16], and Simulation. Several integration methods were also proposed to solve this problem, including AHP and technique for order preference by similarity ideal solution (TOPSIS) [17], AHP and Goal Programming [18], AHP-ELECTRE I [19], Fuzzy AHP-Fuzzy Topsis [20], and Step-wise Weight Assessment Ratio Analysis (SWARA) -Weighted Aggregated Sum-Product Assessment method [21]. Many researchers offer new integration methods because they are considered to have many advantages compared to using a single method.
In previous studies, the weighting of criteria generally used AHP pairwise comparisons. This approach requires a high level of subjectivity in weighting. To reduce this problem, the Rough SWARA approach is proposed to weigh the 3PL provider performance evaluation criteria. The Rough SWARA method and the Complex Proportional Assessment (COPRAS) have not solved the 3PL evaluation problem. The Rough SWARA method was proposed by Zavadskas, et al. [22], which was used for weighting criteria. This method has the advantage of being able to evaluate the ideas of experts and estimate the ratio of relevant interests with the help of rough numbers. This method is used to reduce subjectivity and uncertainty in assessing criteria. Furthermore, the COPRAS method [23] ranks alternative 3PL providers based on their significance and utility level. The contribution of this research is to provide an alternative approach in evaluating the performance of 3PL providers with the Rough SWARA and COPRAS methods.
This paper's structure is presented as follows: Section 2 presents the proposed method of SWARA-COPRAS rough integration and data and case studies. The results of weighting the criteria and performance evaluation of 3PL providers are presented in section 3. Meanwhile, section 4 discusses the conclusions of the research and suggestions for future work.

Proposed Method
This section presents a proposed procedure for evaluating the performance of 3PL providers. This study used Rough SWARA developed by Zavadskas, et al. [22]. The SWARA method allows assessing the opinions of experts on the significance of the criteria and sub-criteria. Many publications have discussed applying an integrated model that involves applying the multi-criteria decision-making method and the rough theory in recent years. Zavadskas, et al. [22] proposed the use of rough theory in order to reduce subjectivity. Furthermore, the COPRAS method is one of the MCDM methods. This method selects the best alternative by considering the positive ideal solution, the negative ideal solution, and the significance of the alternatives considered. This method was developed by Zavadskas and Kaklauskas [23] in evaluating 3PL providers and selecting alternative 3PL providers. The proposed method framework can be seen in Fig. 1.

Fig. 1. A framework of R-SWARA and COPRAS methods
Based on the R-SWARA and COPRAS Method Framework in Fig. 1, the detailed steps that need to be taken to solve the 3PL provider performance evaluation problem are as follows: Step (1) is to establish a set of criteria in evaluating the performance of the 3PL provider. After the 3PL provider, performance evaluation criteria are determined.
Step (2) is that the decision-maker needs to rank the criteria based on priority. The decision maker's ranking criteria are converted into a rough matrix group (Step (3)). The conversion formula can be presented in equation (1) to equation (6). Step (4) is to normalize the matrix ( ) to get the matrix ( ). The criteria that are in the first position have a value of 1. Meanwhile, other criteria can be calculated in equation (7).
Step (5) is to calculate the matrix ( ). The criteria that are in the first position still have a value of 1, while other criteria can be calculated using equation (8).
Step (6) is to determine the weight matrix. The criterion that is in the first position is still worth 1, while other sub-criteria can be calculated using equation (9).
[ , ] Step (7) is to calculate the matrix at the relative weight value ( ) presented in equation (10). Furthermore, Step (8)  [ , ] = [ Step (9) is the decision-maker to assess the 3PL provider. The decision-maker gives the provider rating with a value scale of 0-100. The assessment results for each provider are constructed in a matrix as shown in equation (11). denotes criteria, and j denotes alternatives. In addition, n is the number of criteria, and is the number of alternative providers.
Step (10) is to normalize the decision matrix . The decision normalization formula is presented in equation (12).
Step (11) calculates the normalized weight of the decisionmaking matrix based on equation (13).
Step (12) is to calculate the positive ideal solution ( +) at the criterion value based on equation (14).
Step (13) is to calculate the ideal negative solution ( −) presented in equation (15). Stage (14) is to calculate the relative significance or weight of the relative importance of each alternative presented in equation (16).
Step (15) is to determine alternatives based on the value of relative importance. The formula for determining the value of importance is presented in equation (17). The final step is to determine the performance index calculated based on equation (18).

Case Study
A case study was applied to the largest animal feed company in Indonesia. This study evaluated the providers of raw material shipments from abroad. Three (3) respondents as the decision-makers in this study were the General Manager, Senior Manager, and Supervisor of the import division. Four (4) providers were evaluated in this case study.
The identification of aspects and criteria was based on the results of previous studies conducted by Aguezzoul [24], Hajar and Arifin [16], Bulgurcu and Nakiboglu [25], and Mardani and Saptadi [13], resulting in several criteria in evaluating the performance of 3PL providers. Furthermore, the decision-makers brainstormed the ideas to determine the aspects and criteria used. Six aspects and 16 criteria were successfully collected in assessing 3PL performance. The aspects and criteria used can be seen in Table 1.
Three decision-makers (DM) ranked the criteria and assessed the 3PL provider. The ranking results for each criterion are presented in Table 2. Furthermore, the average 3PL provider rating results from each criterion's decision-makers are presented in Table  3.  Furthermore, the criteria for the duration of invoice submission (C3) was the criterion with the lowest value, amounting to 0.001. These results indicated that the company paid attention to security and safety aspects in the product delivery process.

3PL provider evaluation
From the weighted results, the 3PL provider evaluation was carried out in several stages, such as normalizing the decision matrix, determining the decision weight matrix, determining the useful criteria and useless criteria, and calculating the positive ideal solution and the negative ideal solution. Furthermore, these results can be seen in Table  4 to Table 7. The decision weight matrix Table 5 was based on the multiplication of each value in the decision normalization matrix in each element Table 4 with the weight of each criterion that has been determined using the Rough SWARA method. Table 6 portrays the useful and useless criteria classification where the division was based on whether the criteria were maximum or minimum. The useful criteria were C1, C2, C4, C5, C6, C7, C9, C10, C11, C12, C13, C14, C15, and C16. Meanwhile, the useless criteria were C3 and C8. Table 7 pictures the results of the positive and negative ideal solutions for each provider. For a positive ideal solution, it was obtained from the sum of the decision weights in the useful criteria, while for the negative ideal solution, it was obtained from the sum of the decision weights on the useless criteria for each provider. Meanwhile, for determining provider preference, the relative importance weight ( ) was used as the basis for determining the performance index ( ) for each provider. The results suggested that Provider A was in rank 1, Provider B was in rank 2, Provider C was in rank 4, and Provider D was in rank 3. The performance index value of each provider can be found in Fig. 3.

Conclusion
The purpose of this study was to propose a Rough SWARA and COPRAS procedure in evaluating the performance of 3PL providers. This study observed six aspects with 16 criteria in evaluating the performance of 3PL providers. This study has succeeded in integrating Rough SWARA and COPRAS in the 3PL performance evaluation. The results indicated that product security and safety criteria were the criteria with the most significant weight, followed by the criteria for on-time delivery, responsiveness and flexibility, and financial stability. From the results of the 3PL provider performance evaluation, it showed that provider A had the highest performance index, followed by provider B, provider C, and provider D. These results can be used as a company reference for a description of providers that have a performance index that matches the company's wants and needs. This research can be further developed by considering sustainable aspects such as social, environmental, and economic aspects.