ANALYSIS OF CUSTOMER SATISFACTION IN FREIGHT FORWARDER INDUSTRY USING SERVQUAL, IPA AND FMEA METHODS

A company service’s higher quality is indicated by its service performance’s effectiveness and efficiency. This can increase customers’ satisfaction, thus leads to the increase of company income. This research objective was to know the effects of 5 dimensions constructing service quality, including the effects of service quality on customers’ satisfaction in freight forwarder industry. Current research collected data through purposive sampling, by distributing questionnaire to 86 customers continued by data processing by measuring the gap and tested statistically using SmartPLS program, priority scale mapping using Importance Performance Analysis, and improvement using Failure Mode and Effect Analysis. The first findings obtained from this research was analysis result using IPA discovering 10 indicators which require improvement in which 7 indicators were found the same through GAP measurement. The second finding was the biggest component which needs improvement in freight forwarder service in Indonesia, which is reliability dimension. Therefore, concerns need to be given to these matters so that more accurate and reliable service can be provided.


INTRODUCTION
Economy growth affects logistic business development in Indonesia, indicated by the increase of Indonesian people consumption capacity. However, this matter is not in accordance with the high cost of logistic and low quality of performance. Based on data issued by World Bank in 2018, Logistics Performance Index (LPI) rank of Indonesia was forty six, in which 33% respondents claimed that freight forwarder competence and service quality level in Indonesia was high or higher (Bank, 2018). This value is far from the percentage value of neighboring countries, including Malaysia and Singapore. Based on table 1 above, if benchmarks are provided to several world companies as listed in American Customer Satisfaction Index (ACSI) for consumer shipping, for example UPS in 2018, its customers' satisfaction index was 82% from 100% with company's baseline target was 82%, therefore freight forwarder service satisfaction in Indonesia still requires much improvement.
Based on the previous studies, possible reasons of low service quality of freight forwarder companies were the lack of updated information regarding the applicable regulation in export-import process, lack of understanding related to customs tariffs, failure in goods declaration process, and information delivery to customers which do not pay attention to good grammar (Chatapa, 2017).
Another reason of why local freight forwarder companies service quality is not as good and professional as foreign freight forwarder company is because local companies only compete by decreasing price without adjusting the standard and do not improve its service quality (Truong, 2016).
Gap between the customers' perception and expectation also possibly causes customers' dissatisfaction, for example delay in delivery, lack of understanding related to the documents' preparation so that document errors often occur, discrepancy between agents' policy and applicable regulation, damaged or loss product during delivery, and high cost for customers (Limoubpratum et al., 2020). Therefore, this research carried out investigation whether the five dimensions, including tangible, empathy, responsiveness, reliability and assurance are components constructing service quality as mentioned in previous studies (Kilibarda et al., 2016) as well as to know the effects of service quality on customers' satisfaction according to previous studies (Subhashini & Preetha, 2018).

LITERATURE REVIEW
The text will be written with the normal style with a 5-mm indent on the first line, single line spacing. The font is Times New Roman, 11 pt. Please do not add empty rows before of after chapter titles, unless required (if the title is at the bottom of the page).
According to (Tjiptono, 2019), product quality (goods or service) significantly contributes to customers' satisfaction, customers' retention, wordof-mouth communication, repurchasing, customers' loyalty, market share, and profitability. Research project often employed service quality analysis model of Servqual 5 gaps model which determines the differences between customers' perception and customers' expectation (Zeithaml et al., 1990).
Additionally (Tjiptono, 2019), customers' satisfaction is differences between performance and customers' expectation. Customers' satisfaction indicated that service quality provider must continuously improve their service quality (Ahmed, F , Farooq Jan, M , Ozturk, 2018). Customers' satisfaction is customers' feeling as a result of comparison between performance assessment and customers' expectation over product/service used (Lin et al., 2017).
Freight forwarding is a service related to goods handling, packaging, transporting, storage, consolidation, distribution, and other services and is not limited only to issues related to customs, fiscal, document, and assurance of goods (Watanuki, 2015). Freight forwarding also provides specific service to customers including preparing documents, booking ship/plane ticker, moving goods from the initial point to destination, customs process, information related to applied regulation, insurance process, LC, and others (Grant, 2012).
Structural Equation Modelling (SEM) Partial Least Square (PLS) or SEM-PLS analysis consists of 2 (two) sub-models. The first sub-model is measurement model or outer model which indicates how manifest variable represents latent variable to be measured. The second sub-model is structural model or inner model which shows estimation strength between the latent or construct variables. Furthermore, latent variable formed in SEM-PLS has reflective and formative indicators (Gozhali & Latan, 2015).
Importance Performance Analysis (IPA) method was proposed by (Martilla & James, 1977). Current research also used IPA approach to determine whether the service quality assessed by customers need to be improved (Ding & Tsai, 2012). Other studies also employed IPA in identifying prioritized items that need improvement based on the measurement result and service quality analysis in healthcare (Chang et al., 2019). Therefore, the main purpose of Importance Performance Analysis was to easily identify the attributes based on the interest whether the service has good or bad performance.
Failure Mode and Effect Analysis (FMEA) is a method used to completely identify and comprehend causes and potential failure, as well as failure effects on system or user, for a product or a process; to evaluate risk related to identified failure, effect and cause, as well as to prioritize issues for corrective action; and to identify and carry out corrective action to handle an issue (Carlson, 2012). FMEA is also a technique in the processes of analyzing, defining, and removing potential failure related to process or system before reaching the customers (Andrejić & Kilibarda, 2017).
Based on the above literature review, hypotheses proposed in this research are: Hypothesis (H1): Tangible is one of the components constructing the service quality dimension in the application of freight forwarder industry.
Hypothesis (H2): Empathy is one of the components constructing the service quality dimension in the application of freight forwarder industry.
Hypothesis (H3): Reliability is one of the components constructing the service quality dimension in the application of freight forwarder industry.
Hypothesis (H4): Responsiveness is one of the components constructing the service quality dimension in the application of freight forwarder industry.
Hypothesis (H5): Assurance is one of the components constructing the service quality dimension in the application of freight forwarder industry.
Hypothesis (H6): Service quality has significant effect on customer satisfaction.

METHOD
This research employed purposive sampling technique with respondents involved were 86 people chosen from freight forwarder users data in 2019 in one of subsidiary port service companies engaged in freight forwarder industry in Jakarta. Current research collected data by using questionnaire consisting of general data of respondents and 53 question items using likert scale of 1-5. This then continued by data analysis based on the gap and tested statistically using SmartPLS for its inner and outer model. Furthermore, mapping was also performed by using IPA in addition to improvement on the risks emerged by using FMEA method. The following table presents questions variables, dimensions, and indicators.

Respondents' characteristics
In Based on questionnaire result from 86 respondents involved in the current research, majority of them came from national private companies (54.65%) which engaged as importing companies (39.54%) with the respondents' position level was manager (33.72%), and had been working for the company for 11-15 years (27.91%) as well as having intensity to use freight forwarder service for 11-20 times per month (27.91%). Furthermore, following table shows detail characteristics of the respondents.

Gap analysis
The testing used gap analysis by calculating questionnaire result using Microsoft excel application. Table 4 presents overall gap analysis result for 5 service quality dimensions, with the biggest gap value was reliability dimension of -0.64. The negative value in gap evaluation indicates differences between performance and customers' expectation. Therefore, checking was done on all indicators of 5 service quality dimensions, obtaining 10 indicators in which the highest gap value was.

Outer model analysis
Next stage is processing questionnaire result data using SmartPLS program (Ringle et al., 2015). Outer model measured obtained following results. Figure 2 shows that each indicator has outer loading value of more than 0.7 so that they were valid (Chin, 1998).  Based on table 6 above, composite reliability (CR), cronbach alpha, and rho_A values for all variables obtained above standard value of more than 0.70 (Chin, 1998), thus they met the reliability requirement. Similarly, average variance extracted (AVE) values of all variables also obtained values above standard value of 0.50 (Chin, 1998), thus they met the validity requirements.

Inner model analysis
SmartPls analysis used inner model measurement to know the effects of each variable. The following figure 3 is inner model measurement result.

Fig. 3 Inner model
The following table 7 regarding H1, H2, H3, H4, H5, and H6 hypotheses testing obtained that assurance, empathy, reliability, tangible and responsiveness dimensions construct service quality with 5% significance. Likewise, service quality also significantly affected customers' satisfaction. The testing result can also be seen from t-statistic values of all variables, obtaining value >1.96. Therefore, this result encourages previous studies result from (Le et al., 2019) and (Suprapto & Jani, 2020). This indicates that freight forwarder companies need to seriously concern with five service quality dimensions so that customers' satisfaction is achieved.  (Chin, 1998), thus it was considered as having high/strong predictive accuracy. Likewise in predictive relevance measurement (Q 2 ), tangible, empathy, reliability, responsiveness, assurance and customer satisfaction were also categorized as strong because they have criteria from predictive relevancy measurement if Q 2 value = 0.35 has strong predictive relevance (Chin, 1998). The following table 8 shows measurement results of R-square and predictive relevance. Meanwhile, table 9 presents f-square measurement result, obtaining that service quality has strong effect or good effect towards the five dimensions of tangible, empathy, reliability, responsiveness, and assurance. Likewise, service quality effect on customer satisfaction, obtained 2.992, indicating that it has good or strong effect. The criteria determination in f 2 effect size measurement is f 2 value = 0.02, having small or bad effect size, then if f 2 value = 0.15 has moderate effect size and if f 2 value = 0.35 has strong or good effect size (Chin, 1998).

Importance performance analysis
To see the interest level of each indicator, mapping process was needed by using Importance Performance Analysis (IPA) method which was by comparing performance and expectation of each indicator. IPA analysis process was done by using Microsoft excel application, so that axis x value (performance) obtained 3.8444, while axis y (expectation) obtained 4.4153. furthermore, from axis x and y obtained 10 indicators in quadrant A which is a quadrant considered as important by customers, but companies do not provide good service so that it needs serious attention for improvement. The quadrant mapping is presented in figure 4 below.

Fig. 4 Cartesian diagram
Table 10 explains detail of 10 indicators belong to quadrant A as follow. Mapping result using IPA for measuring risk number 2 which is All workers' competence and knowledge regarding freight forwarder is in accordance with previous research result from (Ding & Tsai, 2012). Then, among 10 improvement prioritized indicators, reliability dimension need much attention for its improvement because there are 3 indicators belong to quadrant A, those are X3.4, X3.5, and X3.6.

Failure mode and effect analysis
The improvement process of 10 mapping result indicators using IPA obtained identification, improvement, and risk assessment of issues using FMEA. Risk assessment of each failure was based on occurrence, severity, and detection level. Table  11 below presents measurement result using FMEA. After the identification and improvement stages, next process is comparing risk priority number (RPN) value before and after improvement. Such comparison is presented in Table 12. The following figure 5 shows significant decrease of RPN value and categories before improvement and after improvement (García et al., 2001). This indicates that improvement is effective to decrease possible risk. Therefore the use of FMEA in this research is accordingly and encourages previous research, that FMEA is needed to identify potential risk from logistic provider (Andrejić & Kilibarda, 2017).

CONCLUSION
Based on results of the current research, it summed up that SmartPLS program successfully proved that tangible, empathy, reliability, responsiveness, and assurance are dimensions constructing service quality in freight forwarder industry. This is indicated by positive path coefficient obtained. Furthermore, service quality also has positive affect on customer satisfaction. In the test using IPA, there were ten indicators belong to quadrant A so that they need prioritized attention to be improved. Based on the IPA measurement on the ten indicators, there were seven same prioritized indicators from the gap measurement result which needs improvement. Thus, identification, problem solving, and risk assessment using FMEA were done on ten of the indicators. Based on FMEA testing, results obtained are significant decrease from RPN value and categories between before and after improvement. Thus shows that improvement efforts are effective in decreasing risk. Furthermore, testing using gap and IPA obtained that the biggest component needs improvement in freight forwarder service in Indonesia, particularly in reliability dimension. Therefore this needs attention so that accurate and reliable service can be provided.