Prioritizing Risks in Production Activities: A Study of Salt Processing Enterprises in the Central Region of Vietnam

Salt is a very important and essential commodity for life [1]. Many archaeological evidences of salt production facilities in ancient Rome show that salt was not only necessary for food preservation but also used in many other production processes. According to Rees [2], before the invention and widespread application of refrigeration in the early 19th and 20th centuries, there were only a few methods that could preserve food, such as drying, smoking, and salting. Of these, salting was the most effective and efficient method. Thus, salt played a fundamental role as a socio-economic driving force of ancient societies and to this day [3-5].

In Vietnam, salt is also an essential commodity, playing an important role in ensuring national food security. Vietnam's salt industry has developed from a long tradition of salt making, with a coastline of 3,260 km from North to South. Vietnam currently has 21 coastal provinces (in the 3 regions of North, Central, and South), but the salt production area is mainly concentrated in the central provinces: Ninh Thuan, Khanh Hoa, Binh Thuan, Ha Tinh, Nghe An, Quang Ngai... Salt products are used as raw materials for a number of industries, such as the chemical industry, seafood processing, food, and pharmaceuticals.

Salt production activities also have the common characteristics of agricultural production activities, always facing many types of risks such as risks of raw materials, weather, coastal environment, food safety, inventory preservation, etc. Therefore, risk identification and prevention are an urgent issue for salt production enterprises in Vietnam in general and salt production enterprises in the Central region in particular to have timely response plans, limiting losses in production activities. However, in reality, enterprises may not have enough resources or cannot respond to all types of risks at the same time. Therefore, the important issue is to analyze, evaluate and prioritize risks to determine their importance so that appropriate risk response attention can be paid [6]. A reasonable and correct approach that managers need to pay attention to at this time is the Analytical Hierarchy Process (AHP) method. AHP is one of the multi-criteria decision-making methods that can assist managers in identifying risks that need to be prioritized for treatment. With AHP, a pairwise comparison matrix is established by the decision maker by comparing the relative importance of criteria to each other according to a separate scale [7].

The objective of this study is to use the AHP method combined with the risk score index and Pareto analysis to determine risk priorities in the salt production activities of salt processing enterprises in the Central region of Vietnam. This study can be useful for salt processing enterprises, helping enterprises identify risks that need to be prioritized and focus on responding to and handling risks to reduce costs, increase profits, and contribute to improving production and business efficiency for enterprises.

3.1 Research design

To prioritize risks in the production activities of salt processing enterprises in the Central region of Vietnam, the AHP method combined with the Risk Score index and Pareto analysis was used in this study. The research procedure is described in Figure 1.

1.png

Figure 1. Research framework

3.2 Qualitative research

3.2.1 Synthesis of studies on risks in production activities

The studies reviewed by the authors include: Indrawati et al. [26]; Van Asselt et al. [27]; Javed et al. [28]; Relkar [29]; Sales et al. [30]; Michalak [31]; Helbig et al. [32]; Nilsen [33]; Nowacki [34].

The preliminary list of 8 risk criteria selected by the authors includes work accident risk, food safety risk, transportation risk, equipment failure risk, inventory management risk, weather risk, raw material supply, and production delay.

3.2.2 Form a team of experts

The expert group consists of 12 members from 6 enterprises specializing in salt production, including 6 directors and 6 factory managers. These are experts with many years of experience and understanding in the field of salt production and processing in Central Vietnam. At the same time, in this expert group, there are also 12 salt workers who have been working in the salt industry for many years.

3.2.3 Expert in-depth interviews

Before conducting the in-depth interviews, the authors sent an email to each expert. The email content asked the experts to consider 8 risk criteria that the authors had preliminarily selected. The in-depth interviews followed.

(1) Objective of in-depth interview: Review risk criteria; which risk criteria need to be eliminated or added? Classify and rank risk criteria.

(2) Participants in in-depth interviews include 12 experts; they are 6 directors and 6 heads of production workshops from 6 enterprises specializing in salt production in the Central region of Vietnam.

(3) Time: In-depth interviews were conducted 3 times. The time for each official discussion and exchange with each expert is from 25 minutes to 40 minutes, not including the time for contacting, responding to comments, and agreeing on opinions.

As a result of the qualitative research, the experts added 3 risk factors to the list of risks, including coastal environmental pollution, preservation risk, and the risk of experienced workers quitting. These 3 risk factors were included in the model by the experts based on the reality at the enterprise. Through active consideration and discussion, all the experts agreed to include these 3 risks in the research model. The experts and the group of authors classified the risks into 3 types, and those are the 3 stages of the entire production process: pre-production, production, and post-production. The hierarchy table and hierarchy model of risk criteria in production activities of salt processing enterprises in the Central region are presented in Table 1 and Figure 2, respectively.

Table 1. Hierarchy of risk criteria in salt production activities

2.png

Figure 2. Hierarchical diagram of risk priority model in salt production activities

3.3 Quantitative research

3.3.1 Structured interviews with experts, collecting comparison scores for each pair of risk criteria

The data collection tool used by the author in the quantitative research was the use of structured interviews with experts using a detailed questionnaire. The questionnaire consists of 3 parts:

Part 1: The author presents a list of risk criteria in salt production activities that were agreed upon during the in-depth interviews.

Part 2: Includes questions comparing the relative importance of each pair of risk criteria. The scale used in the questionnaire for comparing each pair is presented in Table 2.

Part 3: Includes questions collecting data for experts' scores on the likelihood of each risk criterion.

Partial illustrations of the results of this step are shown in Tables 3 and 4.

Table 2. Pairwise comparison scale of importance

Source [35]

Table 3. Illustration of the results of the expert pairwise comparison

Table 4. Illustration of expert risk likelihood scoring results

The experts participating in the structured interview included experts who had participated in the in-depth interview, in addition to 12 salt workers working in the salt industry. To create objectivity in the perception and viewpoint of each expert, the author interviewed each expert independently, in turn, from the first expert to the last expert. By contrasting every risk combination with one another, the goal was to determine the risk criteria importance. It should be noted that the experts must first prioritize the risk criteria according to their feelings and then proceed to compare the importance of each pair of risk criteria.

For collecting data for experts' scoring of the likelihood of each risk criterion. To have a more accurate assessment, closer to reality. With 2 criteria: weather risk (PrPR2) and coastal environmental pollution risk (PrPR3). The author team will collect data from 12 salt workers working in the salt field instead of from 6 directors and 6 factory managers.

3.3.2 Data processing, building comparison matrix

After having collected primary data from interviews with experts, the author proceeded to process the data to build comparison matrices of risk criteria. The principle of data processing is to get the majority of experts' opinions similar in determining which criteria are more important and in assigning points when comparing pairs of risk criteria.

The construction of comparison matrices follows the following principle:

- If the value being evaluated-weighted is to the left of number 1, then record that value exactly in the matrix.

- If the value being evaluated-weighted is to the right of number 1, then record the number equal to the inverse of that evaluation value in the matrix.

The results of the comparison matrix of pairs of risk criteria are shown in Tables 5 to 8.

Level 1 Risk Criteria Comparison Matrix

Table 5. Pairwise comparison matrix of level 1 risk criteria

Comparison matrix of level 2 risk criteria under level 1 risk - Pre-production risk

Table 6. Pairwise comparison matrix of level 2 risk criteria—Pre-production risk

Comparison matrix of level 2 risk criteria under level 1 risk - Risk in production

Table 7. Pairwise comparison matrix of level 2 risk criteria - Risk in production

Comparison matrix of level 2 risk criteria under level 1 risk—Post-production risk

Table 8. Pairwise comparison matrix of level 2 risk criteria—Post-production risk

3.3.3 Calculate the overall weight of each risk criterion and check for consistency

Next, the author will calculate the general weight of the risk criteria. The calculation tool used by the author is Excel software; the steps are as follows:

Step 1: Find the priority vector along with the weights of the level 1 risk criteria (global weight).

Step 2: Find the priority vector along with the weights of the level 2 risk criteria (local weight).

The calculation of finding vectors is performed on each matrix one by one by summing the values of the matrix by column; then each value of the matrix will be divided by the sum of the values of the corresponding column, and the obtained value is substituted for the calculated value. The weight of each criterion (C1, C2, C3, … Cn) will be equal to the average of the values in each horizontal row. The result is a 1-column n-row matrix. The calculated value is only accepted when the consistency ratio CR <= 10% (0.1); if the consistency coefficient is greater than 10%, the expert's comparison result must be checked again [35]. With CI, CR is calculated according to the following formula: Consistency ratio CR: CR = CI/RI.

In which: RI: is the random consistency index (Table 9); CI: is the consistency index CI= (λ max-n)/n-1. λ max: is the eigenvalue of the matrix, $\lambda \max =\sum_{i=1}^n w i \times \sum_{j=1}^n a i j$.

Step 3: Calculate the overall weight.

Table 9. Random consistency index

Source [36]

3.3.4 Calculate the risk score value of each risk criterion

The risk score (RS) will be calculated according to Eq. (1) referenced from Hossen et al. [37].

Risk Score = Impact × Frequency of Occurrence          (1)

Calculating the Impact Level: Once the overall weight has been determined, the overall weight will then be converted into an impact level [37]. Accordingly, the risk factor with the highest overall weight will be assigned the highest impact level of 0.8, and the criterion with the lowest weight will be assigned the smallest value of 0.05. The impact levels of the remaining factors will be calculated in turn using the linear interpolation method. The impact level of the risk will be given according to 5 levels, from 1 to 5, shown in Table 10.

Table 10. Scale of impact level of risks

Source: [38]

Calculating the frequency of occurrence: The results obtained from the questionnaire survey on the likelihood of occurrence of risk factors will be used to calculate the frequency of occurrence.

The frequency of occurrence will be calculated according to Eq. (2) referenced from Hossen et al. [37].

Frequency of occurrence $=\frac{\sum a_i \times n_i}{A \times N}$          (2)

where,

a_i is the weight given to each risk factor (weight from 1 to 5).

n_i is the number of times that weight appears (according to the expert's score).

A is the highest weight = 5.

N is the total number of respondents (number of experts).

The frequency of the risk occurrence will be standardized according to 5 levels, from 1 to 5, shown in Table 11.

3.3.5 Pareto analysis

Pareto analysis is conducted in two main steps. The first step is to identify, classify, and stratify problems, specifically things like errors, defects, delays, etc., and then evaluate and score each problem. Second, it is to display the results in graphic form in a Pareto chart or Pareto diagram so that a few important problems emerge from the general context [24]. In this study, the first main step was carried out by the authors through the implementation of the steps of the AHP method and the calculation of the risk score index. The second main step of Pareto analysis is that the authors draw a Pareto chart; the drawing support tool is Excel software. With a Pareto chart, the line reflects the cumulative percentage of factors that are the principal components. When the cumulative percentage line reaches ≥ 80%, this means that all the factors combined before represent 20% of the causes (i.e., the important few). Remember that the 80/20 rule is a rough guide to typical distributions based on the Pareto principle, so the numbers are not exact, and the total may not add up to 100%. The Pareto chart actually highlights that the weight or impact of the factors that contribute to a particular outcome is not equal [39].

Table 11. Scale of likelihood of risks

Source [38]

5.1 Conclusion

The production activities of salt processing enterprises in Central Vietnam are facing different types of risks. A total of 11 risk criteria were finalized using a systematic literature review with expert input. The authors used a multi-criteria decision-making method, called AHP, to determine the overall weight of the risk criteria, then performed linear interpolation to determine the impact level of the risk criteria, and then calculated the risk score index by calculating the product of the impact level and the frequency of occurrence of the risk criteria. Finally, the authors drew a Pareto chart to find out the priority risks that need to be addressed.

The study's findings demonstrated that the top 5 risk criteria that need to be addressed include weather risk (PrPR2), with a risk score value of 3.5833; storage risk (PoPR1), with a risk score value of 2.6520; food safety risk (RIP2), with a risk score value of 2.4630; coastal environmental pollution risk (PrPR3), with a risk score value of 2.0668; and finally, delayed production risk (RIP5), with a risk score value of 1.7112.

The study has comprehensively synthesized risks in the production activities of salt processing enterprises in the Central region of Vietnam. The study has shown that the AHP method can be applied in combination with the risk score index and Pareto analysis to prioritize risks. The results of this study can also help managers make decisions on choosing priority risk treatment in order to increase the efficiency of the enterprise's production activities.

5.2 Implications

From the above analysis results, with 5 risks that need to be prioritized, the study offers some corresponding management implications to improve the production efficiency of salt processing enterprises in Central Vietnam as follows:

Regarding weather risks: Weather is a natural factor that is very difficult to control. However, if businesses want to avoid this risk, the issue to note is the reserve of raw materials. For the weather in Central Vietnam, storms are very unusual; the prolonged rainy season will affect the salt fields, so it is necessary to monitor the daily weather forecast and take advantage of mild weather to be more favorable in creating raw salt and storing raw salt.

Regarding storage risks: Enterprises need to update new preservation equipment and technology, both saving costs and increasing the efficiency of preservation work such as CAS equipment (Cells Alive System), preservation by irradiation method, etc. Conditions of light and temperature need to be considered when preserving finished products from salt.

Regarding food safety risks: Enterprises producing and processing food need to strictly comply with the regulations of competent authorities on food production, processing, and trading according to the food safety law. The place of food production and processing must ensure that the environment is always clean and dry. The use of additives and preservatives must be in the correct dosage and on the list permitted by the health agency.

Regarding the risk of coastal environmental pollution: Enterprises and salt farmers need to regularly clean the area around the salt fields and check the quality of seawater in the coastal areas approaching the salt fields. In the long term, it is necessary to develop a plan for planning areas, clusters, industrial points, and craft villages specifically for the salt industry.

Regarding the risk of delayed production: Enterprises need to establish a reasonable production plan, strengthen forecasting of demand for Salt and Salt products, and have a plan to reserve raw Salt sources in accordance with market demand.

5.3 Limitations

Regarding the limitations of this study, the experts' evaluation may be biased towards their field of expertise and their level and seniority. Furthermore, with the AHP method, risk analysis is performed using relative priority weights. Therefore, caution should be exercised when constructing a comparison matrix of risk criteria. The study has considered the impact of each risk criterion on the overall production activities but has not considered the interaction of risks with each other, which is also a limitation of the AHP method.

In this study, the author applied the AHP, combined with the Risk Score Index and Pareto analysis, to prioritize risk factors. Future studies by other scholars may combine AHP with one or two other methods to solve risk prioritization problems. The study was conducted at salt production and processing enterprises in Central Vietnam, and the subject was risks in salt production activities, so future studies can be conducted in a different scope or with another research subject, such as risks in rice production activities, risks in steel production activities, risks in cement production activities, etc.