**3. Methodology**

The framework CIRA with eight risk groups was developed and includes the following: supply risks, demand risks, production risks, managemen<sup>t</sup> and operational risks, logistical and infrastructural risks, political risks, policy and regulatory risks and financial risks. An expert evaluation method was employed to assess chosen groups of risks according to their importance. It covers the following four steps: (1) development of a questionnaire; (2) selection of experts; (3) fulfilment of the survey; (4) interpretations of the survey results. Figure 1 presents the process of the research.

**Figure 1.** The process of the research.

The case of Lithuania is used for the research because of its geographical location and size. Furthermore, Lithuania is a small EU country for which trade occurs under all existing inter-lateral agreements with EU countries and other countries. Therefore, it faces all the risks inherent in a small open economy.

According to Libby and Blashfield (1978), seven experts (optimal number) participated in the survey. Table 2 represents qualitative information about the experts. The case of one country (Lithuania) is analysed. Most of the experts were from Lithuania. However, the international experts permitted us to observe the situation from a broader perspective and to have an impartial opinion. Experts filled in the questionnaire for each risk group. A three-level Likert scale was used (low risk, middle risk and high risk).



Three Multicriteria decision support methods were used to assess the analysed risk groups and to obtain the most reliable research results: Simple Additive Weighting (SAW), Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS) and Geometric mean.

SAW method is the most well known and most widely used. It was investigated by MacCrimmon (1968) and is treated as one of the most used multi-criteria decisionmaking methods. This method integrates the values of variables and weights into a single magnitude (Kraujaliene 2019 ˙ ). The application of the SAW method involves three steps: (1) ratios calculation to perform the normalization; (2) calculation of weighted sums of the normalised values; (3) prioritization of risk groups according to the calculated values. All Multicriteria decision methods have their advantages and disadvantages. The SAW method's disadvantage is that all criteria ought to be maximising. In addition, the SAW method requires all the criteria values *rij* to be positive. In our case, all the criteria have positive values and so we did not need to convert them. After receiving the data of expert assessments, the calculation of maximising ratios to perform the normalization was conducted according to the following equation (Gineviˇcius and Podvezko 2008).

$$\mathcal{F}\_{ij} = \frac{r\_{ij}}{\max\_{j} r\_{ij}} \tag{1}$$

The normalization for risk indicators was calculated according to the following equation (Gineviˇcius and Podvezko 2008).

$$\vec{\tau}\_{ij} = \frac{r\_{ij}}{\sum\_{j=1}^{n} r\_{ij}} \tag{2}$$

After the normalization procedure, weighted sums of normalised risk values were calculated according to Equation (3) (Gineviˇcius and Podvezko 2008):

$$S\_{\vec{j}} = \sum\_{i=1}^{m} w\_i \overleftarrow{r}\_{\vec{i}\vec{j}} \tag{3}$$

where:

*wi*—the weight of the *i*th criterion;

*rij*—normalised value from formula (1) and; *m*—number of criteria used for risk evaluation.

Risk groups are ranked according to *Sj*'s calculations. The higher the value of *Sj*, the more important is the risks group.

Hwang and Yoon (1981) introduced the TOPSIS method. The method gain popularity for due to its ease of use and understandable application. Compared to other methods available, TOPSIS may be more stable in the data variation case (Kraujaliene 2019 ˙ ). This method's main principle is that the optimal dote should have the farthest point in the distance from the negative ideal solution point and the shortest line from the positive ideal solution (Dandage et al. 2018). The application of the TOPSIS method involves four steps: (1) normalization procedure; (2) calculation of the best and the worst alternatives; (3) calculation of the distance to the ideal solution and the worst solution; (4) prioritization of risk groups according to the calculated values. TOPSIS can be applied to minimising indicators and maximising ones, i.e., there is no need to convert indicators. The method, TOPSIS, utilizes vector normalization (Podviezko and Podvezko 2014), as described in the following equation.

$$
\widetilde{r}\_{ij} = \frac{r\_{ij}}{\sqrt{\sum\_{j=1}^{n} r\_{ij}^2}} \tag{4}
$$

After the normalization procedure, the best alternative *V*+ and the worst alternative *V*− needs to be chosen.

Then the distance *D*+*j* of every considered alternative to the ideal solution and its distance *<sup>D</sup>*<sup>−</sup>*j* to the worst solution needs to be calculated using the following equation (Podviezko and Podvezko 2014).

$$D\_j^+ = \sqrt{\sum\_{i=1}^m \left(\omega\_i \widetilde{r}\_{ij} - V\_i^+\right)^2} \tag{5}$$

$$D\_j^- = \sqrt{\sum\_{i=1}^m \left(\omega\_i \widetilde{r}\_{ij} - V\_i^-\right)^2} \tag{6}$$

The main cumulative criterion *Cj*'s is calculated (Podviezko and Podvezko 2014) by the following equation.

$$\mathbb{C}\_{j}^{\*} = \frac{D\_{j}^{-}}{D\_{j}^{\*} + D\_{j}^{-}} ; \ (j = 1, 2, \dots, n) , \ \left(0 \le \mathbb{C}\_{j}^{\*} \le 1\right) \tag{7}$$

Risk groups are arranged according to *Cj*'s calculations. The closer the value of *Cj* is to 1, the more important the risk group is.

If the two multicriteria methods results differ in assessing risk groups or possesses the same value, a third method can be used for a more accurate risk group ranking. In the scientific literature, the use of geometric mean weights of (normalised) indicators were considered superior to simpler and more common "weighted arithmetic mean" (Tom and Rogge 2016). The geometric mean is calculated according to Chakraborty and Zavadskas in the following equation (Chakraborty and Zavadskas 2014).

$$
\Pi\_j = \sqrt[m]{\prod\_{i=1}^m \tilde{r}\_{ij}} \tag{8}
$$

See *rij* calculation in Formulas (1) and (2). Coincidence of group values shall be verified before determining the significance of import risks groups by using different multicriteria methods. In the case of discrepancies, the results of different methods are summarised and the final assessment of the significance of risk groups is carried out (Paleviˇcius et al. 2016). The framework of import risk assessment CIRA is based on the results of risk group assessments according to their importance.

#### **4. Research Results**

As mentioned in the literature review, the framework of eight risk groups was developed: supply risks, demand risks, production risks, managemen<sup>t</sup> and operational risks, logistical and infrastructural risks, political risks, policy and regulatory risks and financial risks. The results using the SAW method are presented in Table 3.

**Table 3.** Assessment of import risk groups using the SAW method.


\* Normalised values.

According to the SAW method, the significance of the risk groups is as follows: production risks (the most crucial risk), logistical and infrastructural risks, financial risks,

Political risks

Financial risks

Logistical and

> Policy and regulatory risks

infrastructural

 risks  0.667

 0.333

 0.333

 0.333

managemen<sup>t</sup> and operational risks, political risks, supply risks, policy and regulatory risks and demand risks.

The results using the TOPSIS method are presented in Table 4.


**Table 4.** Assessment of import risk groups using the TOPSIS method.

\* Normalised values.

The best alternative *V*+ and the worst alternative *V*− according to the TOPSIS method are presented in Table 5.

**Table 5.** The best alternative *V*+ and the worst alternative *V*− results according the TOPSIS.


According to TOPSIS expert evaluation method, the significance of the risk according to their importance was as follows: production risks (the most crucial risk), logistical and infrastructural risks, financial risks, supply risks, political risks, managemen<sup>t</sup> and operational risks, policy and regulatory risks, demand risks.

The results of risk group evaluation according to their importance using the SAW and TOPSIS methods differs. In order to determine the straightforward approach of the significance of the risk groups another technique—the Geometric mean (GM) method—is used. The results are presented in Table 6.

> **GM**

 0.607

 0.577

 0.872

 0.662

 0.788

 0.639

 0.630

 0.760

**Risk Group E1 \* E2 \* E3 \* E4 \* E5 \* E6 \* E7 \***Supply risks 0.667 0.667 0.500 0.500 0.500 1.000 0.333Demand risks 0.333 0.667 1.000 0.500 0.500 0.667 0.333Production risks 1.000 1.000 1.000 0.500 1.000 1.000 0.667Managementandoperationalrisks0.3331.0001.0000.5001.0000.6670.333

 1.000

 0.500

 0.500

 1.000

 1.000

 1.000

 0.667

 1.000

**Table 6.** Assessment of import risk groups using the Geometric mean.

> \* Normalised values.

The import risks assessment according to their importance by the Geometric mean are ordered in the following manner: production risks (most crucial risk group), logistical and infrastructural risks, financial risks, managemen<sup>t</sup> and operational risks, political risks, policy and regulatory risks, supply risks and demand risks. The place order of risk groups also differs from previous estimates. The summarised results of risk group assessment are presented in Table 7.

 1.000

 0.500

 0.500

 0.500  1.000

 1.000

 1.000

 1.000  0.333

 0.333

 0.667

 0.667  0.667

 1.000

 0.667

 1.000


**Table 7.** Summarised results of import risk groups.

The importance of import risks summarized by all used methods is as follows: production risks (most crucial risk group), logistical and infrastructural risks, financial risks, managemen<sup>t</sup> and operational risks, political risks, supply risks, policy and regulatory risks and demand risks. According to this assessment, the final framework—CIRA—is developed. This new import risk assessment framework contributes to the systematic approach of a country's international trade risk management.

#### **5. Discussion and Conclusions**

Literature analyses shows that the relevance of the risk is increasing and it covers several aspects. Import risk managemen<sup>t</sup> is important not only for companies but also for each country. Assessing the risks posed by imports is vital for the well-being of the country's population (improving the quality of life) and for its security (in the context of food security, economic and political welfare). It is significant for the country to not only monitor export risks but also to manage import risks. Normally, authors analyse the key risk factors. Our research has shown that the risk factors examined by most authors (Huang et al. 2017; Hyuha et al. 2017) belong to the group of production risks (e.g., country security, unequal distribution of resources and labor market factors), which the country needs to manage the most.

In addition, without managing import risks and especially risks included in production risk group, the country's security is threatened. Leaving it to self-process (under self-interested businesses) may result in insecurity relative to population interests. In order to manage this group of risks, there is a need for political interventions that contribute to OECD (2020) analysis. After analysing the import risk groups presented by various authors, the new framework for CIRA was developed. Our research is primarily based on supply chain risk management, which is also the focus of other researchers (e.g., Nyamah et al. 2017; Spink et al. 2019; Zhao et al. 2020). However, considering the specificities of agricultural products, the role of food quality risk and other import risks observed by other scientists (Welburn et al. 2016; Herrera-Herrera et al. 2019; Attrey 2017; Ruhm 2016; Smith et al. 2017) and that are incorporated into risk groups has been expanded to form a common framework for CIRA. It allows the analysis of all import risk groups of a country by using one framework.

Using multicriteria decision support methods, risk groups were assessed according to the importance of countrywide governance. As all risk groups are significant in the supply chain, it is vital to determine which groups of risks are relevant for governmental management. Since all multicriteria decision support methods have their disadvantages, the use of the three methods ensures an optimal result. In addition, the rating of risk groups allows politicians to focus more clearly, for which risk groups more attention should be given and which should be managed first. It allows using CIRA widely in practice, including the increase in export or reduce of imports and the balance of a country's trade to incorporate import risk management.

The results of our research showed that managing the production risks group is most crucial. This can be explained by the fact that most of the factors involved in this group are related to the primary production of agricultural products and are mainly directed to primary production where the role of the country's governmen<sup>t</sup> could be most significant. Our results show that the import of primary agro products is seen as the most significant risk. However, the situation may differ from one product group to another. For example, the distribution of risk groups in the supply chains of processed food products may vary according to importance.

Further studies are needed to assess the import risks of the different product categories. Nevertheless, managing imports of primary production is the most important for the country. According to our research, the distributions of other risk groups are as follows: logistical and infrastructural risks, financial risks, managemen<sup>t</sup> and operational risks, political risks, supply risks, policy and regulatory risks and demand risks. It demonstrates the importance of supporting sectors managemen<sup>t</sup> in the interest of ensuring the effective functioning of whole supply chains. According to importance, groups of risks can differ in importance due to the countries from which imports are produced. The need for further research is required. It could bring a broader perspective of the importance of the import risks factors and not only risk groups and their effect on business when planning, managing or mitigating an import from different counties or various product groups. Researchers could also analyse import risks in other supply networks (e.g., different retail chains).

Groups of risks can differ according to the countries from which imports are made. The need for further research is required. It could bring a broader perspective of the risk factors and their effect that businesses should consider when planning, managing or mitigating an import from different counties or various product groups.

The research has some limitations. CIRA framework covers risks related at the country level. Future research might cover factors that assesses, with particular attention, and identifies import risk factors for different food product groups. Those factors could also be ranked and compared between different food products groups (e.g., dairy products, grains, beverages, processed food, ready to eat food, etc.). Further research could also bring a wider perspective of the risk factors for separated country groups or different countries. Furthermore, combined (quantitative and qualitative) risk evaluation methods could be used.

**Author Contributions:** Conceptualization, methodology, investigation, writing—original draft preparation L.B.; supervision and writing, review and editing D.J. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Informed consent was obtained from all subjects involved in the study.

**Conflicts of Interest:** The authors declare no conflict of interest.
