The Impact of Inventory Strategy on Operations Performance: Study on Bandung City Traditional Market Trader

ABSTRACT


INTRODUCTION
Traditional markets are part of an urban movement that has mixed activities.Traditional markets serve as a transit space in the exchange of goods and services in a region and develop to cause diverse activity in urban areas.The function of traditional markets in urban areas is fundamental for urban communities and in suburban areas with the aim of meeting daily needs [1].
Traditional markets manifest trade activities between buyers and sellers as a process of interaction between societies to realize well-being.Traditional markets are classified as public facilities in each subdistrict in rural areas or cities.Currently, there are about 13,450 traditional markets throughout Indonesia, which can accommodate 12.6 million traders, and this does not include suppliers and market managers [2].
The market is not just a place where sellers and buyers meet, but it is also a place for social interaction and representation of traditional values.In general, the market has the meaning of a meeting place for sellers and buyers and the functioning of goods or services available for sale which results in the transfer of property rights to potential buyers.According to the Minister of Industry and Trade of the Republic of Indonesia (1998), traditional markets are "markets built and managed by the government, private sector, cooperatives or non-governmental organizations with businesses such as shops, kiosks, and tents, owned and managed by small and medium traders, as well as cooperatives, with small businesses and small capital, by means of the process of buying and selling through bargaining" [3].
Based on the AC Nielsen Survey, the growth of Modern Markets (including Hypermarkets) was 31.4%, while the growth of Traditional Markets -8.1% (SWA, December 2004 Edition).The development of modern retailers has entered the suburbs since the issuance of the trade deregulation policy in 2008.If there is no systematic effort to understand the characteristics and sides of the existence of traditional markets and traders, then the livelihoods of around 12.6 million traditional market traders and their families, employees and commodity suppliers will be threatened with survival [3].
On the other hand, the Covid-19 pandemic is a major event for supply chains in almost all businesses.Nearly eighteen months after the global pandemic was officially announced, many businesses are evaluating the troubling barriers caused by the massive health emergency in different business contexts.Supply chain disruptions are driving businesses to develop capabilities to mitigate the impact of these sources of risk and develop new capabilities have been key to survival during Covid-19.[4] distinguish between different types of supply chain disruptions and the impact of the Covid-19 health crisis, emphasizing how the health crisis impacts global supply chains and forces us to "think in new and foreign ways" [4].
One way for businesspeople to overcome the impact of post-pandemic risks is to create inventory strategies to improve the performance of their business operations.
According to [5] stated that the decisive key in running a business is the rate.Similarly, to ensure that the supply chain is successful, and the fulfillment of orders is right, governance in inventory is needed [6].A study also states that inventory management is needed to balance too much inventory as well as too little inventory [7].Inventory management is also very effective in minimizing inventory turnover while promoting products to the public.Based on theoretical assumptions from [8], [9] shows that a business can adapt by improving operating performance through inventory management strategies [10], [11] Therefore, research on inventory strategies is needed to improve operating performance in traditional markets.The traditional market has the most complicated types of inventories in terms of perishability of raw materials, semi-finished or finished products.However, there hasn't been much research focused on traditional market traders.This research will support the continuity of traders in traditional markets, especially in the city of Bandung.So that traditional market traders can survive during post-pandemic.

Conceptual Model
At this stage, as much information is collected as possible related to all trader activities, with the aim of knowing the real condition of the object to be studied, namely the Bandung City Traditional Market.The conditions obtained from field studies are expected to be quite detailed and complete, so that they can be used in formulating variables with clear specifications.
After going through the field study stage, it is then necessary to formulate variables, dimensions and indicator items that will be used in solving existing problems.The inventory strategy approach used in this study includes variable Strategic Supplier Partnership (SSP), Economic Order Quantity (EOQ), and Just-in Time (JIT).The formulation of variables was carried out by referring to the research of [12], researchers used a research model to examine Operations Performance (OP) in traditional market traders in the city of Bandung with inventory strategy approach which can be seen in Figure 1.

Hypothesis Determination
Figure 1 shows a new synthesis model to be tested by researchers.From Figure 1 the researcher identified the following hypotheses: • H1: The Economic Order Quantity  Data collection is carried out through questionnaires to test the suggested hypotheses of determinants that have been encoded and have reliable decisions.The operational variables on the questionnaire must be encoded in precisely defined terms (Table .1) The questionnaire is divided into three stages.At the first stage, the researcher clarifies the relationship between the synthesis model of the research proposal and the measurement scale used in individuals.All items in the questionnaire were measured using 5 Likert scales starting where 1 = strongly disagree, 2 = disagree, 3 = neutral, 4 = agree, and 5 = strongly agree described in Appendix A. Research design consists of four main determinants of Economic Order Quantity (EOQ), Just-in Time (JIT), Strategic Supplier Partnership (SSP) and Operations Performance (OP) which the researcher forms and incorporates into the research model.
In the second stage, an experiment was carried out through the distribution of questionnaires to all traditional market traders in the city of Bandung using a google form.The population of respondents in this study was all traders in the Bandung City Traditional Market.Based on the selection of samples will use purposive sampling, namely withdrawal with certain criteria.The criteria for drawing this sample are Traditional Market traders in the city of Bandung who sell food raw materials.The number of samples to be used in this study is based on Hair Jr et al., (2021) who explained that the minimum sample size based on the minimum R2 values starts from 0.1, 0.25, 0.5 and 0.75 in the endogenous construct in SEM for significance levels of 1%, 5% and 10% by looking at the maximum number of constructs in the PLS Path Model.Based on the number of free variables of this study in the SEM size there are 3 with R2 0.25 and a significance level of 5%, the minimum number of samples is 59.

Variable Statistical Analysis
Variable statistical analysis was carried out to provide an overview of the research questionnaire data filled out by respondents.This analysis is carried out by categorizing the average value per question indicator (mean), standard deviation value, excess kurtosis value, skewness value and the average of each variable (total mean).
The variables above show that the effect of Just-in Time (JIT), Strategic Supplier Partnership (SSP), and Economic Order Quantity (EOQ on Operating Performance (OP) on traditional market traders in the city of Bandung is largely dominated by affirmative answers to questions asked by researchers.
It is also known for the highest index in the DSA1 statement, namely "We usually sell every day" where this item corresponds to the behavior of traditional market traders who sell every day.
As for the lowest index in the PP1 statement, namely "Sharing information related to our goods to customers through information technology (WhatsApp, social media, etc.)".This is appropriate because traders in traditional markets rarely provide information to customers through information technology because some customers come directly to the traditional market without being given information first by market traders.

Outer Model Testing
Outer models are used to evaluate measurement models consisting of the reliability and validity of the models proposed by researchers.The test is measured using SmartPLS 4.0 by measuring a set of relationships between latent variables and their indicators [14].

Convergent Validity Testing
To analyze reflective models, outer loading greater than 0.6 is recommended by [14].However, if the outer loading is less than 0.4, the reflective indicator should be removed.When outer loading between 0.4 and 0.7 it is recommended to keep or delete items depending on the outer load (height) of other items [14].Based on this theory, researchers took a minimum value of 0.4.
Based on Figure 2 and Table 3, all measuring items have met the outer loading value testing requirements so that they can be said to be valid and can be used to measure each of the latent variables.

Discriminant Validity Testing
The next process carried out by this study is to measure discriminant validity to measure each variable in the conceptual model of the study.The validity of this discriminant wants to know the correlation between constants in the model referred to as cross loadings [15].
Existing Based on Table 4, it shows that all cross-loading values on each intended construct are greater than the cross-loading values with other constructs.It can be concluded that all indicators are valid and there are no problems with discriminant validity.

Construct Reliability Testing
The Reliability Test is used to measure the reliability of each construct.This test can be known through the reliability of the composite as well as the alpha Cronbach [16] Typically, during reliability analysis, consistency of determinants is obtained when Cronbach's alpha coefficient (α) is at 0.5 or more reflecting acceptable determinants [17] and composite reliability (CR) is accepted when Cronbach's alpha exceeds the minimum acceptable limit of 0.70 ( [18]; [17].
Table 5 shows that the results of construct reliability tests show that all latent variable values have Cronbach's alpha values at 0.5 or more reflecting acceptable determinants and composite reliability exceeding the minimum acceptable limit of ≥ 0.70.Thus, the construct is acceptable in reliability.

Inner Model Testing
After the estimated model meets the criteria of the measuring model (outer model), the next structural model (inner model) testing is carried out.According to [19] the evaluation of structural models (inner models) aims to predict relationships between latent variables.[14] in [20] suggest looking at the value of the coefficient of determination (R 2 ), the value of effect size (f 2 ), and the fit model to assess the structural (inner model).

Coefficient of Determination Testing (R-Square)
The R-Square test is used to show how large an exogenous variable is in describing its endogen variable.The smaller the R-square value the smaller the ability of exogenous variables to explain endogenous variables.
In this study, there was one endogenous variable, namely Operations Performance (OP) which was influenced by 3 exogenous variables, namely Economic Order Quantity (EOQ), Just-in Time (JIT), and Strategic Supplier Partnership (SSP).
From Table 6 above, R 2 the Operations Performance (OP) construct is 0.629.These results show that Operations Performance (OP) variables can be explained by exogenous variables, namely Economic Order Quantity (EOQ), Just-in Time (JIT), and Strategic Supplier Partnership (SSP) of 62.9% while the rest is explained by other exogenous variables outside this study.

Cohen Effect Testing (f-square)
The f 2 test is known as the simultaneous test or Anova test, which is a test to see how all its free variables affect together on their bound variables.The effect size according to Cohen (1988) is small (f 2>0.02), medium (f 2>0.15), and large (f 2 >0.35).
Based on the test results in Table 7, it can be found that Economic Order Quantity (EOQ) has a small influence on Operations Performance (OP), Just-in Time (JIT) has a major influence on Operations Performance (OP), and Strategic Supplier Partnership (SSP) has a small influence on Operations Performance (OP).

Model Fit Testing
Testing the fit model in this study was carried out using two testing models, including standardized root mean square residual (SRMR) and normal fit index (NFI) proposed by [21] that the model will be considered to have good fit if the value of the standardized root mean square residual (SRMR) is below 1.00 [22].Another conformity index is the normed fit index (NFI) with the calculation of the value of Chi2 [23] The Chi-square value is then compared with the given benchmark in the context of Goodness of Fit.Referring to [23], acceptable conformity values when using Chi-square as a measurement are greater than 0.9 (Chi2> 0.9).
Based on Table 8, the results showed that the model in this study had a good fit because it had a standardized root mean square residual (SRMR) value below 1.00 and a Chi-square value of >0.9.

Hypothesis Testing
The value of the path coefficient that is in the range of values -1 to +1, where the value of the path coefficient close to +1 represents a strong positive relationship and the value of the path coefficient that is -1 indicates a strong negative relationship.Tstatistics aims to determine the value of significancy between variables in the study.The hypothesis is accepted if the T-statistical value is greater than 1.96 whereas if the Tstatistical value is less than 1.96 then the hypothesis is rejected or the null hypothesis is accepted ( [14] in [20] Based on Figure 3, Table 7, and

Discussion
This study aims to study and analyze the effect of inventory strategies on operating performance in traditional market traders in the city of Bandung.
Based on the test results on Just-in Time (JIT) against Operations Performance (OP) shows the existence of a T-Statistical value of 6,203 (>1.96), an f-square value of 0.552 and a p-value of 0.000 (<0.05) so that it can be concluded that the second hypothesis (H2) is accepted where Just-in Time (JIT) positively and significantly improves Operations Performance (OP).
Meanwhile, based on the test results on Economic Order Quantity (EOQ) and Strategic Supplier Partnership (SSP) on Operations Performance (OP) showed that there were T-Statistic values of 1,638 (<1.96) and 1,305 (<1.96), f-square values of 0.042 and 0.031 respectively and p-value values of 0.101 (>0.05) and 0.192 (>0.05) so that it can be concluded that the first (H1) and third (H3) hypotheses were rejected where the Economic Order Quantity (EOQ) and Strategic Supplier Partnership (SSP) does not affect Operations Performance (OP).
The results of this study are not in line with the research conducted by [12] which found that the Just-in Time (JIT) inventory strategy was found not to affect Operations Performance (OP) activities, while in this study it was found to have a very large influence on the operating performance of traditional market traders because according to researchers Just-in Time is the most suitable inventory management technique used in traditional traders due to the JIT technique is a means to meet customer demands with minimum delays.The JIT technique not only can control the right items, in the right quantity, at the right time, but also you can take those supplies to the right place (Muller, M., 2019).The research of [24] [25] and [26] also proves that JIT strategies are effective for reducing inventory costs and unnecessary shortages.Therefore, JIT is considered a valuable strategy for inventory management.This is in accordance with the principles carried out by traditional traders where traditional traders usually sell goods that are often needed by their customers with quantity, quality and prices that suit their customers.Meanwhile, EOQ and SSP do not affect the performance of traders' operations in traditional markets because EOQ is usually done to evaluate activities in a year to plan optimal orders that need to be carried out in the future, while traditional market traders rarely carry out such plans and calculate the goods for a year at a minimum cost for a year as well.Likewise, SSP is rarely used because traders in traditional markets rarely enter into goods sale and purchase agreements with their suppliers because if the supplier does not produce, traditional market traders tend to have many suppliers without agreements to overcome their inventory shortages.

(
EOQ) strategy significantly and positively improves Operations Performance (OP) in traditional market traders.• H2: Just-in Time (JIT) strategy significantly and positively improves OP Operations Performance (OP) in traditional market traders.• H3: Strategic Supplier Partnership (SSP) significantly and positively improves Operations Performance (OP) in traditional market traders.

Figure 1 .
Figure 1.Conceptual Research Model 2.3 Data Collection and Model ExperimentsData collection is carried out through questionnaires to test the suggested hypotheses of determinants that have been encoded and have reliable decisions.The operational variables on the questionnaire must be encoded in precisely defined terms (Table.1)Thequestionnaire is divided into three stages.At the first stage, the researcher clarifies the relationship between the synthesis model of the research proposal and

Figure 3 .
Figure 3. Bootstrapping Test Results Source: Smartpls 4.0 Output Results (2022) tables or figures are presented with sufficient explanations and by including numbers and titles.Complete the existing tables and figures by writing the source under each table/figure.The table is created without a vertical border.Example table.

Table 5 .
Construct Reliability Test Results

Table 6 .
Coefficient of Determination Test Results (R-Square)