ReseaRch PaPeR Journal of Agricultural and Marine Sciences 2022, 27(1): 99–111 DOI: 10.53541/jams.vol27iss1pp99-111 Received 17 May 2021 Accepted 08 Oct 2021 Risks in the Sri Lankan Banana Supply Chain: Analysis through an Interpretive Structural Modeling Shanaka Rajakaruna1,*, Alge Wattage Wijeratne2 Shanaka Rajakaruna1,*( ) shanaka.rajakaruna@gmail.com, 1Dalian Maritime University, Transportation Engineering College, No:1 Ling- hai Road, Dalian, China, 2Department of Agribusiness Management, Sabaragamuwa University of Sri Lanaka Introduction Banana (Musa acuminata L.) is one of the favourite fruit crops grown in Sri Lanka covering an area of 72,010 hectares with an annual average produc- tion of 864,120 tons (AgStat, 2019). It is also known as one of the staple foods that helps the food security of many tropical countries (Ghag & Ganapathi, 2017). Out of the 1000 varieties available around the world, Sri Lan- ka grows 29 of these varieties (Ratnasinghe, 2003). These are mainly divided into four main groups namely; Sour, Seeni, Kolikuttu and Anmalu (AgStat, 2018). Sri Lanka is geographically located in a favourable position to grow banana. Additionally, the soil and other climatic condi- tions are also useful to the growth of the crop (Perming, 2013). Due to these facts, there is a potential to increase the harvest and income of farmers. However, irrigation problems, weather patterns, cost of inputs, pests and diseases, political impacts, and climate changes, are seen as the crucial impediments to planting and growing ba- nana in Sri Lanka (Perming, 2013). There is always a struggle among farmers to continue agriculture due to the low income. Due to the short- er yield time, low cost of replanting and low overheads, more farmers grow banana as perennial crop (Arvan- itoyannis and Mavromatis, 2009). Although, there is a املخاطر يف سلسلة توريد املوز السريالنكي: التحليل من خالل النمذجة اهليكلية التفسريية شاانكا راجاكاروان1* , اجلي وااتج وجيرياتين Abstract. Although banana is an important fruit in Sri Lanka that gives a good income to farmers. Researchers have in- dicated that the majority of the crop is damaged from the farm gate to the consumer table. Accordingly, this research was addressed ‘to find the risks of the banana supply chain in Sri Lanka. The Interpretive Structural Modeling (ISM) approach was used in this research to find the risks in the banana supply chain of Sri Lanka. According to the literature survey and data analysis, several risks such as human and personal, production, financial, land prices, biological and environmental, information, political, weather-related, management and operational, agricultural policy, logistics and infrastructure, country’s economic cycle, price or market and input risks can be encountered in the banana supply chain. These risks were categorized into different levels according to the influence they provide to the supply chain. Identifying these risks on a hierarchical model was significant to the administrators, managers and farmers to minimize them effectively to manage the supply chain. The analysis in this study further indicated that production, financial, biological and environmental, weather-related events, political situations, and logistics and infrastructure were the most significant risks. Dealing with these risks can minimize the effect of other risks, improve the income of the farmers and maximize consumer choice. Keywords: Agriculture, Banana, Interpretive Structural Modeling (ISM), Risks, Sri Lanka, Supply Chain. امللخص:علــى الرغــم مــن أن املــوز يعتــر فاكهــة مهمــة يف ســريانكا حيــث يوفــر دخــًا جيــًدا للمزارعــن، إال أن الباحثــن أشــاروا إىل أن غالبيــة احملصــول يبــدأ ابلتلــف منــذ خروجــه مــن بوابــة املزرعــة وحــى طاولــة املســتهلك. وفًقــا لذلــك، تنــاول هــذا البحــث خماطــر سلســلة توريــد املــوز يف ســريانكا. يف هــذا البحث مت اســتخدام هنج النمذجة اهليكلية التفســريية )ISM( للعثور على املخاطر يف سلســلة توريد املوز يف ســريانكا. وفًقا للمصادر العلمية وحتليل البيــاانت، هنــاك العديــد مــن املخاطــر مثــل املخاطــر البشــرية والشــخصية واإلنتاجيــة واملاليــة وأســعار األراضــي واملخاطــر البيولوجيــة والبيئيــة واملعلوماتيــة والسياســية واملخاطــر املتعلقــة ابلطقــس واإلدارة والتشــغيل والسياســة الزراعيــة واللوجســتيات والبنيــة التحتيــة والــدورة االقتصاديــة للبلــد و خماطــر األســعار أو الســوق كل هــذه املدخــات ميكــن مواجهتــا يف سلســلة توريــد املــوز. مت تصنيــف هــذه املخاطــر إىل مســتوايت خمتلفــة وفًقــا للتأثــري الــذي حتدثــه يف سلســلة التوريــد. إن حتديــد هــذه املخاطــر علــى منــوذج هرمــي ســيكون مهًمــا للمســؤولن واملديريــن واملزارعــن لتقليــل هــذه املخاطــر بشــكل فعــال وإلدارة سلســلة التوريــد. وقــد أشــارت نتائــج هــذه الدراســة إىل أن اإلنتــاج، واألحــداث املاليــة والبيولوجيــة والبيئيــة، واألحــداث املتعلقــة ابلطقــس، واألوضــاع السياســية، واخلدمــات اللوجســتية والبنيــة التحتيــة كانــت مــن أهــم املخاطــر. إن التعامــل مــع هــذه املخاطــر قــد يــؤدي إىل تقليــل أتثــري املخاطــر األخــرى، وحتســن دخــل املزارعــن وزايدة خيــارات املســتهلك. الكلمات املفتاحية: الزراعة ، املوز ، النمذجة اهليكلية التفسريية )ISM( ، املخاطر ، سريانكا ، سلسلة التوريد. 100 SQU Journal of Agricultural and Marine Sciences, 2022, Volume 27, Issue 1 Risks in the Sri Lankan Banana Supply Chain: Analysis through an Interpretive Structural Modeling relatively higher turnaround of income from planting banana, full benefit of the yield is not received by farm- ers due to various problems in the supply chain. The yield of Sri Lankan banana has increased with- out a major increase in the harvested area. Although the prices of the various crops of banana have increased from 20 to 30 per cent from 2016 to 2017, the wholesale prices have only increased from 10 to 20 per cent (Ag- Stat, 2018) during the same period. This indicates that there is a decrease of profit from farmer to wholesaler. Researchers have indicated that the state of the insub- stantial and non-standard transportation methods from each point in the supply chain have made a significant quantity of the crop being perished during transporta- tion (Wasala et al., 2014). The supply chain always encourages a system-wide approach that helps identify the commonalities and the differences in separate segments (King and Venturini, 2014). As described by Jaffee, et al. (2010), banana sup- ply chain in Sri Lanka also covers various activities, such as raw materials with equipment and technology; pro- duction methodology adopted for harvesting. In addi- tion, post harvesting methods such as re-plantation and crop rotation; processing the harvest by diversifying the product range; promoting the product by carrying out marketing campaigns; and distribution of the products throughout the country to the consumers affected the supply chain. There is a physical financial information flow occurs in this supply chain as in the case of others (Partane et al., 2020). The major attraction of growing banana to the farm- ers is the economic gain. A shorter life cycle and ease of planting also contribute to the growth (Kudagam- age et al., 2002). However, 30% of the crop is damaged during the journey from the ‘farm to fork’ (Ekanayake and Bandara, 2002). Researchers indicated that improp- er handling, bad packaging, and excess use of pesti- cides, mechanical damage and various other problems throughout the supply chain as the reasons for this loss (Hathurusinghe et al., 2012). Additionally, the involve- ment of middle vendor by double handling the product contributes to these post-harvest losses and reduces the level of profits of farmers. Although there was no in- crease in the cultivated land, crop yield has increased between 2014 and 2017 (IndexMundi, 2020; AgStat, 2018). Due to the various factors, the advantage of pro- duction growth has not reached farmers (Ratnasinghe, 2003). Additionally, the relationship between income and profitability depends on the cost of cultivation. Such a decreasing trend in profitability may have various asso- ciated reasons. Having an effective supply will leverage the firms and improve profitability (Salvia, 2020). This research, therefore, was focused on identifying the risks that affect the banana supply chain to propose strategies to minimize such risks that would contribute to increase the profit of the sector. Many programs have been launched by the governmen- tal and non-governmental sector to reduce the negative effects on banana supply chains. There has been previ- ous research conducted to find the effectiveness of the banana industry (Ekanayake and Bandara, 2002), factors affecting the transport chain (Abeysekera & Abeysekera, 2008), methods of increasing the yield (Ekanayake and Bandara, 2002), supply chain issues (Sarananda, 1999) and packaging (Wilson, 1996) of banana in Sri Lanka. The risk will restrict the efficient function of the sup- ply chains (Crouhy et al., 2006). A large number of risks have been studied by previous researchers concerning agriculture supply chains (Jaffee et al., 2010). Managing these risks is also necessary to increase efficiency and effectiveness. However, in this research, we consid- ered the factors with a holistic view of the supply chain. Although different types of risks are agreed upon by different researchers, they may differ from situa- tion, sector, time and type of operation (Crouhy et al., 2006). Risk management is also a prominent factor in agriculture (Jaffee et al., 2010). Although the concept of risk management in agriculture is less-known in Sri Lanka, it is being practiced in other regions. Risk man- agement in the agriculture supply chain can also be explained as “managing the most demanding events” (Jaffee, et al., 2010) that negatively affect the financial and information flows in the supply chain. It is difficult to mitigate all the risks to the lowest level. However, trying to minimize such risks will positively affect the collective impact of the whole supply chain (Hathurus- inghe et al., 2012). It did not mention what risks are the most significant in banana supply chains. A holis- tic approach to agricultural supply chain risks is not deemed practical due to the diverse nature of each crop and the other aspects such as region and country that may affect the supply chain (Dolgui, et al., 2020). The situation is the same with Sri Lankan banana industry. Thus, the main objective of this study was to iden- tify the risks in the banana supply chain of Sri Lan- ka and rank them according to their severity. These findings will enable the regulators to mitigate them depending on the impact of each risk. We used Inter- pretive Structural Modeling (ISM) to analyze the data. Subsequently, this research will enable administrators and farmers to minimize the likelihood and the conse- quences of risks in the banana industry. This will en- able a smoother efficient flow of product, finances and information. This is the intial study carried out to find the risks in the banana supply chain in Sri Lanka. Ad- ditionally, initial study to apply ISM in the Sri Lankan banana industry. Thus, this research can open a new paradigm in the agriculture supply chain of Sri Lanka. Methodology In this study, firstly we researched journal articles to find various risks in the banana industry as published by 101Research Paper Rajakaruna, Wijeratne the previous researchers. Number of journal sites were used for this search. Accordingly, we found 312 articles with the term ‘agriculture supply chain’ and ‘risks’ in the publication title. Thereafter, this search was further re- fined to find the words ‘agriculture’, ‘risk’ and ‘banana in- dustry’ in the abstract and/or key words of each article; 21 articles qualified the criteria. Having gone through the articles, it was revealed that 8 of those 21 articles do not contain any direct relevance to agricultural risk in the banana industry. Finally, we established 16 risks that have a relationship to ‘agriculture risks’ and ‘ba- nana industry’ and thirteen articles was finally selected. Secondly, we discussed the existence of these 16 risks in the Sri Lankan banana industry with a randomly se- lected sample of government agriculture officers, farm- ers and vendors. The sample consisted of 10 government agriculture officers, 30 small, medium and large scale farmers and 15 vendors. We requested the sample to rank the risks according to their knowledge and expe- rience. The sample represented all banana growing dis- tricts in Sri Lanka. Thirdly, we introduced Interpretive Structural Mod- eling (ISM) as the principal method of data analysis in this research. Accordingly, we provided these initially agreed 16 risks to a focus group consisted of six academ- ics and administrators in the field of agriculture. Con- sidering the socio-economic conditions as well as with relevance to the banana industry in Sri Lanka, the fo- cus group agreed on fourteen most relevant risks that are suitable to the Sri Lankan banana supply chain. The focus group decided to drop Marketing Risk since it is already including in the Price or Market Risk. Addi- tionally, Institutional Risk was also dropped since it is already a part of Management & Operational Risk. Con- sequently, the following fourteen risks were considered for the ISM analysis considering their relevance to the Sri Lankan economic and agricultural situation: (i) Hu- man & Personal Risk, (ii) Production Risk, (iii) Financial Risks, (iv) Risk of Increase In Land Prices, (v) Biological & Environmental Risk, (vi) Information Risk, (vii) Polit- ical Risk, (viii) Weather-Related Risks, (ix) Management & Operational Risk, (x) Agricultural Policy Risk, (xi) Lo- gistics & Infrastructure Risk, (xii) Country’s Economic Cycle, (xiii) Price or Market Risk, and (xiv) Input Risk. We use ISM to study the hierarchy of these risks. This method enables to convert poorly expressed mod- els to more detailed and defined models that can be fur- ther tested with the use of data. ISM can also be used for many other purposes, such as finding the interrela- tionships among variables, to identify and rank the vari- ables, ease of educating the public and critically analyse the results for improvements (Sushil, 2012) However, the interrelationship of the factors can be considered as the most important (Attri et al., 2013). The method is interpretive since the judgment of a group decides the relationship among different variables. Modelling meth- odology in ISM facilitates to develop a diagraph that showed the direction and order in the variables (Attri et al., 2013). ISM has several advantages as an analysis tool. All possible pairwise elements are selected through a literature survey was further purified by the respon- dents’ recommendations. The use of such transitive in- ferences can reduce the required rational queries mak- ing the process efficient (Sushil, 2012). ISM also helped to record complex issues efficiently and systematically. The structural model generated through ISM graph- ically presented a complex problem. This graphical representation could depict the problem where inter- disciplinary and interpersonal communication can be effectively. Thenceforth, ISM served as a tool to devel- op a deeper understanding when identifying, analyzing and taking actions to each element (Kumar et al., 2018). Every effort was made to reduce the negative effects of ISM in this research. As having a larger number of vari- ables may increase the complexity of methodology, this research was only considered the most significant risks factors in the banana supply chain with the assistance of expert opinion. Discussions were carried out among the researchers and the focus group on the driving and the dependence power effect variables. ISM was used across many research fields such as information technology (Pfohl et al., 2011), supply chain (Khan and Rahman, 2017), agriculture (Gardas et al., 2017), and quality man- agement (Muruganantham et al., 2018). Risk identifica- tions and analysis was the core of these research. Thus, it was proved that ISM was a realistic analysis tool in identifying risks. Although agricultural risks was identi- fied in previous research (Gardas et al., 2017; Sudarshan et al., 2013), there is a deficiency of literature on banana supply chains, particularly in the Asian region. This re- search deemed significant considering factors, such as the income of banana farmers, nutritious value in the food, improvement of export income and diversification of the agricultural sector in Sri Lanka. Steps were tak- en according to the ISM methodology when developing and identifying the risks to the banana supply chain. Step One - Listing of Variables Most of the research published in various sources was discussed and the variables that relate to the banana supply chain were indicated in Table 1. Step Two - Establishing a Contextual Relation- ship among Variables Contextual and pairwise relationship among variables were identified in this step. The consultation and assis- tance of an expert panel consisted of researchers, aca- demics and practitioners were used initially to nominate the variables and to find the pairwise relationship among them whilst removing the transitive links. As indicated previously, the opinion of experts from various spheres in the agricultural field was used to identify the rela- tionship of these risks to one another. Afterwards, this appropriate relationship and significance were further 102 SQU Journal of Agricultural and Marine Sciences, 2022, Volume 27, Issue 1 Risks in the Sri Lankan Banana Supply Chain: Analysis through an Interpretive Structural Modeling examined. The relationship among variables was shown in the SSIM as below: V= i will achieve j. A= j will achieve i. X= i and j will achieve each other. O= i and j are unrelated. where i and j are variables in the banana supply chain and Ψ indicates the relationship among the variables. Structural Self Interaction Matrix (SSIM) is a compar- ison that was based on pairwise relationships. Thus, SSIM was obtained by asking a question of whether ‘variable i’ influenced the ‘variable j’. This was described in Figure 1. Step Three - Structural Self Interaction Matrix (SSIM) Development Table 1. Structural self-interaction matrix (SSIM) of the causal factors of Risks in Banana Supply Chain: Source Research data Variable j Risk Parameter 14 13 12 11 10 9 8 7 6 5 4 3 2 1 Variable i 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Human & Personal Risk Production Risk Financial Risks Risk of Increase in Land Prices Biological and Environmental Risk Information Risk Political Risk Weather-Related Risks Management & Operational Risk Agricultural Policy Risk Logistics & Infrastructure Risk Country’s Economic Cycle Price or Market Risk Input Risk X V V X V A V V A V V A A - V V V V V A V V A V O A - V V V V V V V V A V V - A A O A O A O O O A - A V V A V A V V O - O V V V V X V V - O V V A X A O - O V V A O O - V O V V V - A V V A - X O V - A A - A - - Note. Where the driver is V= element i will help to achieve element j Where the driver is A= element j will help to achieve element i Where the driver is X=element i and j will achieve each other Where the driver is O= i and j are unrelated Accordingly, the decision of the expert panel was considered to develop the SSIM in table 2. Figure 1. Structural Self Interaction Matrix (SSIM). 103Research Paper Rajakaruna, Wijeratne SSIM was established for all variables to identify the pairwise relationship for the research area of the banana supply chain of Sri Lanka. Step Four - Initial Binary Reachability Matrix (IBRM) Development The previously created SSIM reachability matrix was used to form the IRM, and this was also further anal- ysed for transitivity. Transitivity was an assumption that defined the relationship among factors such as if ‘A’ was related to ‘B’ and also ‘B’ was related to ‘C’ then we hy- pothesized that ‘A’ was also related to ‘C’. The SSIM de- veloped in the previous step was converted to a binary matrix. The V, A, X, and O was converted to 1’s and 0’s. IBRM is presented in Table 2. The 1’s and 0’s were re- placed according to the following rule: • If i,j entry is V then it becomes a 1 and the correspond- ing j, i entry be a 0. • If i, j entry is A then it becomes a 0 and the correspond- ing j, i entry be a1. • If i, j entry is X then it becomes a 1 and the correspond- ing j, i entry also be a 1. • If i, j entry is O then it becomes a 0 and the correspond- ing j, i entry also be a 0. Step Five- Developing a Final Reachability Ma- trix (FRM) After considering the transitivity as indicated in Step Four, FRM was developed. The FRM was developed by considering transitivity. Accordingly, the reachability matrix’s transitivity was analysed with equations 1 and 2 according to Huang et al. (2005): i. RM = SSIM+U1 ii. RM*= RMX= RMX+1 +1,X> 12 According to the above equation, U is the unit of matrix and X indicates the power. Also RM* is the reachabili- ty matrix. Boolean multiplication and addition system was used (Lin et al., 2010) in determining the transitivity values. This suggests us: 1.1=1;1+1= 1;1.0= 0;1+0= 0+1= 1;1.0 = 0.1 = 0. Step Six - Carrying out Level Partitions Level partitions were checked to identify the rank of each element. Reachability and the antecedent sets were used to identify such hierarchy. Reachability and an- tecedent sets were found from the FRM as used to de- velop the level partitions. These assisted to identify the elements in a ranked pattern (Suhil, 2012). This can be expressed as: R(Di)= R(Di) ∩ A(Di) Accordingly, Di is element and all the other elements that help to attain, R(Di) is reachability set with the el- ement as well as all other elements that help to attain it, and A(Di) is antecedent set that consists of the subjec- tive element and all other elements to achieve it. Step Seven – Creating a Diagraph A direct graph (diagraph) was created using the final reachability to present the relationship among variables. Previously, different levels of drivers in the banana sup- ply chain were identified to develop a diagraph. Step Eight - MICMAC Analysis Impact Matrix Cross-Reference Multiplication Applied to a Classification (MICMAC) analysis was performed to find the driving and dependence power of each variable. Table 2. Initial Binary Reachability Matrix: Parameter Variable j 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Variable i 1 2 3 4 5 6 7 8 9 10 11 12 13 14 1 1 1 1 1 0 0 0 0 1 1 0 0 1 0 1 1 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 1 0 1 1 1 0 1 1 0 1 1 0 0 1 0 1 1 0 1 0 0 1 0 0 0 0 0 0 1 0 1 1 1 1 0 1 1 1 1 0 1 1 0 1 1 0 0 0 1 0 0 0 0 0 0 0 0 1 1 0 1 0 0 1 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 0 0 1 1 1 0 1 1 0 1 0 1 1 0 1 1 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 1 1 0 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 0 1 1 0 1 0 0 1 1 1 1 1 1 1 0 1 1 0 1 1 0 0 1 Source – Survey data 104 SQU Journal of Agricultural and Marine Sciences, 2022, Volume 27, Issue 1 Risks in the Sri Lankan Banana Supply Chain: Analysis through an Interpretive Structural Modeling Results and Discussion As described previously, several risks related to the ba- nana supply chain from the ‘farm to fork’ was explored in this research. Identifying such risks can enable us to undertake the most vital risk that can have a trickledown effect to minimize other related risks. Such action can, therefore, reduce the cost and effort of tackling each risk independently. This can create a cost and benefit advan- tage for the farmers and vendors as well as the govern- ment to attain an effective banana supply chain in Sri Lanka. All such risks researched previously were selected by the expert panel and these were used in this research. Types of Risk Risk of Country’s Economic Cycle Domestic crop production can be greatly influenced the economic cycle and governmental policies to promote agriculture. Research has showed that the economic cri- sis can positively affect agriculture since more people can turn to agriculture due to increasing unemployment (Ogboru et al., 2018). Moreover, the elasticity variance of agriculture was low during different economic peri- ods from a recession to an expansion (Contre and Gol- din, 1990). However, the sustainability of agriculture has a concern on economic, social and ecological sustain- ability (Ghag and Ganapathi, 2017). Price or Market Risk: Due to the nature of demand and supply pattern in agriculture, there is always a risk involved in market price. However, the price or market risk can vary from product to product (Rosch, 2018). Banana being a short-term crop and can help the subsis- tence farmers in Sri Lanka, since there is always a price variation. From 2015 - Jan 2019 monthly banana prices have increased by 65% (IndexMundi, 2020). This did not consider the monthly variations that were almost 15% in the year 2017 (for Ambul only) (CBSL, 2018). Risk of Increase in Land Price: Land prices in the Co- Figure 2. ISM model for Banana Supply Chain Risks. 105Research Paper Rajakaruna, Wijeratne lombo district have increased by 18% compared to 2017 (CBSL, 2019). Central Bank (2019), report further indi- cates that ‘three sub-indices of Land Price Index, namely ‘residential, commercial and industrial have contributed’ to this increase. It can be claimed that this has a trick- le-down effect on the increase in land prices in other dis- tricts. Growth in population and improvement of roads and infrastructure has also contributed to this de-urban- ization (Rosch, 2019: Ogboru, et al., 2018). As a result, harvesting of banana did not increase from 2014 to 2017 (AgStat, 2018). Input Risk: Banana farmers face a shortage of ap- proved tissue cultured nursery plants due to minimum research carried out in Sri Lanka (Perming, 2013). The situation about the labor, pesticides and the special planting and yielding equipment were also the same (Hanson et al., 2004). Researchers indicated that 2% of the yield was damaged due to bad handling at the farm gate since the right equipment was not available (Wasala et al., 2014). Moreover, most of the banana plantations on the Island were organic (Sarananda, 1999), thus re- quired knowledge and the inputs to develop popular or- ganic farming were also in short supply. Biological and Environmental Risk: Researchers have classified biological and environmental risks in agricul- ture to two areas namely, dependent and independent from the business activities (Zakharchenko, 2017). The growth of agriculture including farming and animal husbandry harms the environment (Jaffee et al., 2010). Heavy use of fertilizer can deteriorate the quality of the topsoil and deplete the groundwater sources (Zakharch- enko, 2017). Since the farmers lack knowledge on sci- entifically grounded crop rotation methods, problem of soil and groundwater contamination has worsened. Also, the continuously deteriorating environment due to deforestation and extreme climatic conditions can be negatively contributed. Although, there was much less use of fertilizer in the banana plantations in Sri Lanka (Sarananda, 1999), other factors did not affect biological and environmental risk. Logistics and Infrastructure Risk Climacteric and the perishable nature of banana make them highly vulnerable to postharvest losses (Wasala et al., 2014). The available transportation and storage con- ditions can also deteriorate the situation further. As in- dicated in Figure 2, banana supply chain in Sri Lanka has many transportation loops and storage points. This can make it vulnerable to more postharvest losses during transit and it was calculated that this loss amounts to 30% of the yield (Ekanayake and Bandara, 2002). Political Risk: To develop an efficient and resilient agriculture mechanism in any country, innovative ex- perimentation and research are needed (Bennett et al., 2014). Also, there should be a ‘holding hands’ approach between the farmers and the authorities to develop a winning supply chain. Since all governmental issues are politically driven, it is difficult to have the expected sus- tainability in agriculture (Ghag and Ganapathi, 2017). This has a direct impact on the agriculture policy risk that can be discussed next. However, other factors are influenced by political risks, such as land availability, supply chain effectivity and agricultural decentralization (Dolgui et al., 2020). These can be influenced by the pre- vailing political regime. Such short-termism approach of the politicians had negatively influenced the agriculture industry including the banana supply chain (Contre and Goldin, 1990). Considering the influence that the politi- cal risk can bring to the banana supply chain (Ekanayake and Bandara, 2002), it can be included as a major risk. Agricultural Policy Risk: As described previously, the government agricultural mechanism can waste re- sources and efforts in practices that are ineffective and does not provide adequate opportunities for innovation (De Silva and Kawasaki, 2018). As a reason, the farming community is unable to reap the benefits of such govern- mental efforts. Presently, it is provided by the local ag- ricultural advisers and this does not have an end to end process where a crop can be looked after by the govern- mental mechanism from seed to yield (Abeysekera and Abeysekera, 2008). Additionally, the policy of various governments changed time to time ad these could also slow down the growth. Weather-related Risk: Agriculture contributes 7.5% of the total GDP of Sri Lanka in the year 2018 (CBSL, 2019). That said, there is a correlation between weather-relat- ed activities and agriculture productivity (De Silva and Kawasaki, 2018). Natural disasters are increasing due to climate change. Therefore, whenever there is a climatic change, there is a risk of the banana crop being damaged. Financial Risk: Managing the costs incurred through planting to consumption, availability of various financ- ing methods, and the knowledge of the farmers to manage the cash flows are known as the highest risks in agriculture (Komarek et al., 2020). There is a lack of knowledge in cash flow management throughout the farming communities of Sri Lanka (Kiriveldeniya and Rosairo, 2020) Since there is a shortage of awareness in sound financial control due to the illiteracy of the farm- ing communities (Wanigasundera and Atapattu, 2019), objective nature of farming could not be achieved. Whilst the fact that the role of the government to im- prove the financial acumen of the farmers are not being negated, there is still much the farming communities can also do in reducing the financial risk by way of coopera- tives and other cartels (Kiriveldeniya and Rosairo, 2020). Production Risk: Production risk involves all the risk factors that help production from inputs, process, weather, biological and environmental factors (Shahzad and Abdulai, 2019). Although some of these risks were discussed previously, we highlighted production risk also as a separate risk phenomenon due to the signifi- cance. Since there are other contributory factors such as pests, disease and machinery efficiency and these did 106 SQU Journal of Agricultural and Marine Sciences, 2022, Volume 27, Issue 1 Risks in the Sri Lankan Banana Supply Chain: Analysis through an Interpretive Structural Modeling not include previously. The food and Agriculture Organ- isation (FAO) of the United Nations have claimed that there is “20 to 40 percent of crops that are lost due to pest” (FAO, 2020). These pest problems have affected the food security of various countries. Although there are no statistical data, this must-have affected the Sri Lankan Agriculture sector too. The recent locust at- tacks on the plantations in Kurunagala District in Sri Lanka can be one of the examples (Shivantha, 2020). Human and Personal Risk: There are various hu- man and personnel risks that the agriculture indus- try faces every day in Sri Lanka (CBSL, 2019). Among them a scarcity of labor force (Karunagoda, 2004), in- adequate health and safety systems (Weerasinghe et al., 2020), non-effective training to farmers including the partners of the banana supply chain (Kiriveldeniya and Rosairo, 2020), non-availability of pension schemes (Perera and Weerakkody, 2018) and ill health among farming communities (DCS, 2017) are the most aggra- vating factors that affect human and personal health. Operational Risk: Operational risk is defined as something that might happen and its effect(s) on the achievement of objectives (ISO, 2018). When consid- ering the banana supply chain, there may be numerous objectives including growth, expansion profitability and resilience. The Ministry of Agriculture has laid down policy objectives and road maps for the whole industry. However, there is a lack of proactive planning to have a first and second-line defense model to minimize or to eradicate the consequences and likelihoods of prob- lems that may encounter when reaching these strate- gic objectives (Nisansala et al., 2020). Additionally, the changes in the policy and political environment can also affect the drive towards the set strategic objective. Information Risk: Information has become a signifi- cant requirement in our everyday life. Thenceforth, it is also vital to the development of the banana supply chain (Bo et al., 2014). Accordingly, the right mandate and Figure 3. MICMAC Analysis Diagram, Banana Supply Chain Risks. 107Research Paper Rajakaruna, Wijeratne infrastructure were set by the authorities to have a free flow of supply chain-wide information availability. The banana supply chain can achieve many advantages by effectively employing costly inputs (Dolgui et al., 2020). Even though, the government has taken several steps there are still gaps in the system. Lack of knowledge and the ability of the farmers, lack of information sources, unreceptive attitude of the authorities and slow flow of information are some problems, and these have to be re- solved to have a sound information regime (Gardas et al., 2017). Considering the impact that information has on the development of the banana supply chain, information risk can be judged as risk as identified in this research. The final Binary Reachability Matrix is presented in Table 3. According to the results, financial risks takes the highest rank of the set of drivers followed by pro- duction risks. This implies that policy-level initiatives must be in place to reduce both financial and produc- tion risks as a priority. The third rank is taken by bio- logical and environmental risks that should be handled separately. When the dependence power is examined the top rank is taken by price or market risks. This sig- nals that price regulation and access to market price and market place timely may help reduce such dependency. Models The results of Level Partitions are presented in Table 4. Thus, the element which has the highest number of in- tersections and antecedents were considered as the first level partition. This application was continued until all variables were removed. These partitions were then used to develop the diagraph and the ISM model at the later stage. Thenceforth, the FRM achieved is shown in Table 4 indicating the driving and dependence power as well as the transitivity of the banana supply chain risks in Sri Lanka. This further indicates the role of financial risks and logistic and infrastructure risks in the overall con- text of the banana supply chain in Sri Lanka. Results of diagraph and formation of ISM-based model shows the graphical representation of the rela- tionships of the elements according to different hierar- chy. According to the diagraph, it can be seen that “Lo- gistics & Infrastructure risk (Driver 11)” and “Financial risks (Driver 3)” can be considered as the most signifi- cant risks to affect the others in the banana supply chain. Therefore, these risks need to be dealt with at the outset. However, management and operational risk are at the top of the diagraph. This indicates that treating this risk can greatly influence all the other risks. Accordingly, the final model of the diagraph after removing the transitiv- ity is presented in Figure 2. Results of MICMAC analysis helps to understand the driving and dependence power of the risks in the banana supply chain. There are four groups in this analysis. Namely, autonomous, dependent, linkage and indepen- dent. These were also numbered as region one to four respectively in Figure 3 (Kumar et al., 2018). There were no risks in the banana supply chain that falls into the autonomous region which is also called region one. The dependence cluster which is named as region two had Information risk (6),” “country’s economic cy- cle (12),” “price or market risk (13),” and “management & operational risk (9)”. These risks had high dependence and low driving power. The region three (linkage) had, “human & personal risk (1),” “risk of an increase in land prices (4),” “input risk (14),” and “agriculture policy risk (10)”. This indicates that action taken to minimize these risks can affect other risks. Risks such as “production risk (2),” “financial risk (3),” “biological & environmental risk (5),” “weather-related risks (8),” “political risk (7),” and “logistics infrastructure Table 3. Final Binary Reachability Matrix: Source – Survey data. Drivers Variable j 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Driving Power Rank 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Human & Personal Risk Production Risk Financial Risks Risk of Increase in Land Prices Biological & Environmental Risk Information Risk Political Risk Weather- Related Risks Management & Operational Risk Agricultural Policy Risk Logistics & Infrastructure Risk Country’s Economic Cycle Price or Market Risk Input Risk Dependence Power Rank 1 1 1 1 1 0 0 1# 0 1 1 0 0 1 9 V 0 1 1 0 0 0 0 0 0 0 1 0 0 0 3 IX 0 0 1 0 0 0 0 0 0 0 0 0 0 0 1 X 1 1# 1 1 1 0 1 1 0 1 1 0 0 1 10 IV 0 1 1 0 1 0 0 1 0 0 0 0 0 0 4 VIII 1 1# 1 1 1 1 0 1 1 1 1 0 1 1 12 II 0 1 1 0 0 0 1 0 0 0 0 0 0 0 3 X 0 1 1 0 1 0 0 1 0 0 1# 0 0 0 5 VII 1# 1 1 1 1 1 1 1 1 1# 0 1 1 1 13 I 0 1 1 0 1 0 1 1 0 1 1 0 0 0 7 VI 0 0 0 0 0 0 1 0 0 0 1 0 0 0 1 X 1 1 1 1 1 1 1 1 0 1 1 1 1 1 13 I 1 1 1 1 1 0 1 1 0 1 1# 0 1 1 11 III 1 1 1 1 1 0 0 1 0 1 1 0 0 1 10 IV 7 12 13 7 10 3 7 10 2 8 10 2 4 7 V II I V III VII V III VIII IV III VIII VI V Note. 1# entries indicate the transitivity 108 SQU Journal of Agricultural and Marine Sciences, 2022, Volume 27, Issue 1 Risks in the Sri Lankan Banana Supply Chain: Analysis through an Interpretive Structural Modeling Table 4. Drivers to Risks in Banana Supply Chain Partition. Drivers(Di) Reachability R(Di) Antecedent A(Di) Intersection R(Di) ∩ A(Di) Level I 1 2 3 4 5 6 7 8 9 10 11 12 13 14 1,4,6,9,12,14 1,2,4,5,6,7,8,9,10,12,13,14 1,2,3,4,5,6,7,8,9,10,12,13,14 1,4,6,9,12,13,14 1,4,5,6,8,9,10,12,13,14 6,9,12 4,7,9,10,12,13,14 1,4,5,6,8,9,10,12,13,14 6,9 1,4,6,9,10,12,13,14 1,2,4,6,8,10,11,12,13,14 9,12 6,9,12,13 1,4,6,9,12,13,14 1,2,3,4,5,8,10,11,14 2,3,11 3 1,2,3,4,5,7,8,10,11,14 2,3,5,8 1,2,3,4,5,6,8,9,10,11,13,14 2,3,5,7 2,3,5,8,11 1,2,3,4,5,6,7,8,9,10,13,14 2,3,5,7,8,10,11 7,11 1,2,3,4,5,6,7,8,10,11,12,13,14 1,2,3,4,5,7,8,10,11,13,14 1,2,3,4,5,7,8,10,11,14 1,4,14 2 3 1,4,14 5,8 6,9 7 5,8 6,9 10 11 12 13 1,4,14 Level II 1 2 3 4 5 6 7 8 10 11 12 13 14 1,4,6,13,14 1,2,4,5,6,7,8,10,13,14 1,2,3,4,5,6,7,8,10,13,14 1,4,6,13,14 1,4,5,6,8,10,13,14 6 4,7,10,13,14 1,4,5,6,8,10,13,14 1,4,6,10,13,14 1,2,4,6,8,10,11,13,14 12 6,13 1,4,6,13,14 1,2,3,4,5,8,10,11,14 2,3,11 3 1,2,3,4,5,7,8,10,11,14 2,3,5,8 1,2,3,4,5,6,8,9,10,11,13,14 2,3,5,7 2,3,5,8,11 2,3,5,7,8,10,11 7,11 1,2,3,4,5,6,7,8,10,11,12,13,14 1,2,3,4,5,7,8,10,11,13,14 1,2,3,4,5,7,8,10,11,14 1,4,14 2 3 1,4,14 5,8 6 7 5,8 10 11 12 13 1,4,14 Level III 1 2 3 4 5 7 8 10 11 13 14 1,4,13,14 1,2,4,5,7,8,10,13,14 1,2,3,4,5,7,8,10,13,14 1,4,13,14 1,4,5,8,10,13,14 4,7,10,13,14 1,4,5,8,10,13,14 1,4,10,13,14 1,2,4,8,10,11,13,14 13 1,4,13,14 1,2,3,4,5,8,10,11,14 2,3,11 3 1,2,3,4,5,7,8,10,11,14 2,3,5,8 2,3,5,7 2,3,5,8,11 2,3,5,7,8,10,11 7,11 1,2,3,4,5,7,8,10,11,13,14 1,2,3,4,5,7,8,10,11,14 1,4,14 2 3 1,4,14 5,8 7 5,8 10 11 13 1,4,14 Level IV 1 2 3 4 5 7 8 10 11 14 1,4,14 1,2,4,5,7,8,10,14 1,2,3,4,5,7,8,10,14 1,4,14 1,4,5,8,10,14 4,7,10,14 1,4,5,8,10,14 1,4,10,14 1,2,4,8,10,11,14 1,4,14 1,2,3,4,5,8,10,11,14 2,3,11 3 1,2,3,4,5,7,8,10,11,14 2,3,5,8 2,3,5,7 2,3,5,8,11 2,3,5,7,8,10,11 7,11 1,2,3,4,5,7,8,10,11,14 1,4,14 2 3 1,4,14 5,8 7 5,8 10 11 1,4,14 109Research Paper Rajakaruna, Wijeratne risk (11)” fell into region four (also known as an inde- pendent). These risks therefore considered as the most significant risks in the banana supply chain since they drive the other risks. This phenomenon was also evident in the ISM model in Figure 2 where these risks were in the lower layers of the risk model. We consider that this would help the administra- tors to take necessary measures to minimize the risks in the banana supply chain. It shall be emphasized that administrators, practitioners and farmers must take the required action to reduce these risks to improve the ba- nana supply chain and increase the profit as well as the benefits to the customers. Also, the structural model we introduced can show the hierarchy of action needed to improve the performances of the banana by undertak- ing each risk considering the interrelationship among each risk as shown in Figure 3. Since there were no risks that fall into ‘region one’, none of the risks shown in the research can be disconnected. Therefore, the authori- ties should look into all risks that are identified from a holistic point of view. Additionally, the risks that are in ‘region three’ are high independence and driver power, therefore minimizing these risks can reduce the effect on them as well as the other risks such as the ones in ‘region two’ which can be influenced by them. Addition- ally, the risks that are in ‘region four’ of the MICMAC analysis diagram lie on the lower part of the ISM mod- el (Figure 3). Therefore, minimizing such as production risk (2), logistics & infrastructure risks (11) and the fi- nancial risk (3), affect the whole banana supply chain. Moreover, administrators and managers can minimize these risks as the basis of all other risks. This research explored the risks that are in the Sri Lank- an banana supply chain. However, there were some lim- itations encountered during the research due to time and financial restraints. Accordingly, we could not com- pare the results of ISM against other available models such as Structural Equation Modeling (SEM), Pareto analysis or Failure Mode and Effects Analysis (FMEA). Additionally, there may be other risks that affect the ba- nana supply chain. We call for future research in these areas. Although this research was only towards the ba- nana supply chain, this model could be applied to other areas in agriculture in Sri Lanka as well as in other re- gions. Finally, we hope that the productive potential of the banana supply chain can be increased by minimizing these risks. Conclusion This study was initiated to find the risks that are avail- able in the Sri Lankan banana supply chain. Accordingly, various agricultural risks were found through a literature survey and these were forwarded to a randomly selected sample of farmers, vendors and government-appoint- ed agricultural officers. These represented all banana growing districts of Sri Lanka. Once the sample agree- ment was obtained, these risks were presented to a fo- cus group to carry out the pairwise relationship among them whilst removing the transitive links. Afterwards, the data were analyzed through ISM according to their driving and dependence power. It was revealed that four risks are more significant than others. Minimizing or eradicating these four risks can influence other risks. Level V 2 3 5 7 8 10 11 2,5,7,8,10 2,3,5,7,8,10 5,8,10 7,10 5,8,10 10 2,8,10,11 2,3,11 3 2,3,5,8 2,3,5,7 2,3,5,8,11 2,3,5,7,8,10,11 7,11 2 3 5,8 7 5,8 10 11 Level VI 2 3 5 7 8 11 2,5,7,8 2,3,5,7,8 5,8 7 5,8 2,8,11 2,3,11 3 2,3,5,8 2,3,5,7 2,3,5,8,11 7,11 2 3 5,8 7 5,8 11 Level VII 2 3 11 2 2,3 2,11 2,3,11 3 7,11 2 3 11 Level VIII 3 11 3 11 3 7,11 3 11 110 SQU Journal of Agricultural and Marine Sciences, 2022, Volume 27, Issue 1 Risks in the Sri Lankan Banana Supply Chain: Analysis through an Interpretive Structural Modeling The administrators can take necessary action to discuss a methodology to minimize them throughout the sup- ply chain. This study is significant to Sri Lankan banana supply chain since there is a 30% wastage of the crop from ‘farm to fork’. This was the initial study carried out in Sri Lanka regarding agriculture using ISM methodol- ogy. This study can also be used in other countries in the region with the necessary adaptations. References Abeysekera T, Abeysekera S. (2008). Alternative supply chain management practices and the performance of marketing channels in fresh fruit and vegetable market- ing in Sri Lanka, http://www.fao.org (accessed 27 Octo- ber 2019). AgStat. 2018. Centre, Socio-Economics and Planning, Peradeniya - Kandy Sri Lanka: Department of Agricul- ture, https://www.doa.gov.lk (accessed 12 April 2020). AgStat. 2019. Centre, Socio-Economics and Planning, Pera- deniya - Kandy Sri Lanka: Department of Agriculture, https://www.doa.gov.lk (accessed 28 February 2021). Arvanitoyannis I, Mavromatis, A. (2009). Banana cultivars, cultivation practices, and physicochemical properties. Critical Review in Food Science and Nutrition 49: 113-135. Attri R, Dev N, Sharma V. (2013). Interpretive structural modelling (ISM) approach: An overview. Research Jour- nal of Management Sciences 2(2): 3-8. Bennett E, Carpenter S, Gordon J, Ramankut N. (2014). To- ward a More Resilient Agriculture. The Solutions Jour- nal 5(5): 65-75. Bo Y, Ping S, De-long D. (2014). Risk Assessment and Con- trol of Agricultural Supply Chain on the Internet of Things. Journal of Industrial Engineering and Engineer- ing Management 2014(03): 10-23. CBSL. 2018. Monthly Price Bulletin, Colombo: Central Bank of Sri Lanka, https://www.cbsl.gov.lk (accessed 10 June 2020). CBSL. 2019. Annual Report, Colombo: Central Bank of Sri Lanka, https://www.cbsl.gov.lk (accessed 3 may 2020). Contre F, Goldin I. (1990). Agriculture and The Economic Cycle: An Economic and Econometric Analysis with Spe- cial Reference to Brazil, Brasília: OECD Development Centre. https://www.oecd.org (accessed 3 May 2020). Crouhy M, Galai D, Mark R. (2006). The Essentials of Risk Management. McGraw Hill, New York. DCS (2017). Agricultural Household Survey, Colombo: Department of Census and Statistics Sri Lanka, http:// www.statistics.gov.lk (accessed 3 may 2020). De Silva M, Kawasaki A. (2018). Socioeconomic vulnera- bility to disaster risk: A case study of flood and drought impact in a rural Sri Lankan community. Ecological Eco- nomics 152: 131-140. Dolgui A, Ivanov D, Rozhkov M. (2020). Does the ripple effect influence the bullwhip effect? An integrated anal- ysis of structural and operational dynamics in the sup- ply chain. International Journal of Production Research 58(5): 1285-1301. Ekanayake S, Bandara L. (2002). Development of banana leather. Annals of the Sri Lanka Department of Agricul- ture 4: 353-358. FAO, 2020. Food and Agriculture Organization, http:// www.fao.org (accessed 15 June 2020]. Gardas BB, Raut RD, Narkhede B. (2017). Modelling causal factors of post-harvesting losses in vegetable and fruit supply chain: An Indian perspective. Renew- able and Sustainable Energy Reviews 80: 1355-1371. Ghag S, Ganapathi T. (2017). Genetically modified bananas: To mitigate food security concerns. Scientia Horticul- ture 214(1): 91-98. Hanson J, Dismukes R, Chambers W, Greene C, Kremen A. (2004). Risk and risk management in organic agricul- ture: Views of organic farmers. Renewable Agriculture and Food Systems 19(4): 218–227. Hathurusinghe C, Vidanapathirana R, Rambukwella R, So- maratne T. (2012). A Study on Value Chain of Pineapple and Banana in Sri Lanka. (accessed 4 October 2018). Huang J, Tzeng G, Ong C. (2005). Multidimensional data in multidimensional scaling using the analytic network process. Pattern Recognition Letters 26(6): 755–767. IndexMundi. (2020). IndexMundi. https://www.indexmun- di.com (accessed 9 June 2020). ISO (2018). ISO 31000. Geneva, Switzerland: BSI Standards Limited. Jaffee S, Siegel P, Andrews C. (2010). Rapid Agricultural Supply Chain Risk Assessment: A Conceptual Frame- work. The World Bank, World Bank. http://documents1. worldbank.org (accessed 6 July 2020). Karunagoda K. (2004). Changes in labor market and do- mestic agriculture. Sri Lankan Journal of Agricultural Economics 6(1): 83-97. Khan I, Rahman Z. (2017). Brand experience anatomy in hotels: An interpretive structural modeling Approach. Cornell Hospitality Quarterly 58(2): 165-178. King R, Venturini L. (2014). Demand for quality drives changes in food supply chains, Chicago: Economic Re- search Service/USDA, Chicago (accessed 20 July 2020). Kiriveldeniya K, Rosario H. (2020). Value chain actors, farm-gate price and farmer loyalty in strategic vertical coordination in the maize out-grower farming in Sri Lanka. The Journal of Agricultural Sciences - Sri Lanka 15(2): 154-172. Komarek A, Pinto A, Smith V. (2020). A review of types of risks in agriculture: What we know and what we need to know. Agricultural Systems 178: 1-18. Kudagamage C, Chandrasiri G, Razmy A. (2002). Analysis of long-term trends in the banana sector of Sri Lanka, 111Research Paper Rajakaruna, Wijeratne Annals of Sri Lanka. Annals of the Sri Lanka Depart- ment of Agriculture 4: 21-32. Kumar P, Ahmed F, Singh R, Sinha P. (2018). Determination of hierarchical relationships among sustainable devel- opment goals using interpretive structural modelling. Environment, Development and Sustainability 20(5): 2119–2137. Lin Y, Lin C, Yu H, Tzeng G. (2010). A novel hybrid MCDM approach for outsourcing vendor selection: A case study for a semiconductor company in Taiwan. Expert Sys- tems with Applications 37(7): 4796-4804. Muruganantham G, Vinodh S, Arun CS, Ramesh K. (2018). Application of interpretive structural modelling for an- alysing barriers to total quality management practices implementation in the automotive sector. Total Quality Management & Business Excellence 29(5-6): 524-545. Nisansala W, Abeysingha N, Islam A, Bandara A. (2020). Recent rainfall trend over Sri Lanka 1987–2017. Inter- national Journal of Climatology 40(7): 3417-3435. Ogboru I, Abdulmalik F, Park I. (2018). Government expen- diture on agriculture and its impact on unemployment reduction in Nigeria: 1999 –2015. International Journal of Economics, Commerce and Management 6(3): 1-25. Partane J, Kohtamäki M, Patel P, Parida, V. (2020). Supply chain ambidexterity and manufacturing SME perfor- mance: The moderating roles of network capability and strategic information flow. International Journal of Pro- duction Economics 221: 1-10. Perera E, Weerakkody W. (2018). Determinants of Post-Re- tirement Employment in Sri Lanka. Kelaniya Journal of Management 7(2): 38-53. Perming E. (2013). Paddy and banana cultivation in Sri Lanka - A study analysing the farmers’ constraints in agriculture with a focus on Sooriyawewa D.S. division, Lund Sweden: Department of Physical Geography and Ecosystem Science Physical Geography and Ecosystems Analysis Lund University Sölvegatan. Pfohl HC, Gallus P, Thomas D. (2011). Interpretive struc- tural modelling of supply chain risks. International Jour- nal of Physical Distribution & Logistics Management 41(9): 839-859. Ratnasinghe G. (2003). Present status and future research need on banana nematodes in Sri Lanka. In Train- ing Workshop on Enhancing Capacity for Nematode Management on Small-scale Banana Cropping Sys- tems, Laguna, Philippines: INIBAP-AP, Los Baños. Rosch S. (2018). United States Department of Agricul- ture-Economic Research Service. https://www.ers.usda. gov (accessed 9 October 2018). Rosch S. (2019). United States Department of Agriculture Economic Research Service. https://www.ers.usda.gov (accessed 8 June 2020). Salvia L. (2020). The restructuring of Italian agriculture and its impact upon capital-labour relations: Labour con- tracting and exploitation in the fresh fruit and vegetable supply chain of the Lazio Region, Central Italy. Agrarian Change 20(1): 98-112. Sarananda K. (1999). Country paper - Sri Lanka, Report of the APO Seminar on Appropriate Postharvest Tech- nologies for Horticultural Crops in Asia. Bangkok-Thai- land, 230-236. Shahzad M, Abdulai A. (2019). Production risk manage- ment in agriculture and farm performance in rural Paki- stan: Role of adaptation to climate change. Agricultural & Applied Economics Association Annual Meeting, At- lanta, GA. Shivantha K. (2020). Sri Lanka Mirror. https://srilankamir- ror.com (accessed 13 July 2020). Sudarshan GM, Anand B, Sudulaimuttu MB. (2013). Mar- keting & post-harvest losses in fruits: Its implications on availability and economy-A study on pomegranate in Karnataka. International Journal of Management and Social Sciences 2(7): 34-43. Sushil M. (2012). Interpreting the interpretive structural model. Global Journal of Flexible Systems Management 13(2): 87-106. Wanigasundera W, Atapattu N. (2019). Extension reforms in Sri Lanka: lessons and policy options. In: Extension reforms in Sri Lanka: lessons and policy options. Babu S, Joshi P. eds. Academic Press, Washington DC, p. 79-98. Wasala W, Dharmasena D, Disanayaka T. (2012). Physical characteristics and mechanical properties three com- mercially grown banana cultivars in Sri Lanka. Tropical Agricultural Research 24(1): 42-53. Wasala W. (2014). Postharvest losses, current issues and de- mand for postharvest technologies for loss management in the main banana supply chains in Sri Lanka. Journal of Postharvest Technology 2(1): 80-87. Weerasinghe W, Rajapaksha E, Gunawardena W, Samara- kone T. (2020). Relationship between management prac- tices and calf welfare in mid-country dairy farms in Sri Lanka. Tropical Agricultural Research 31(1): 103-113. Wilson N. (1996). Supply chain management: a case study of a dedicated supply chain for bananas in the UK gro- cery market. Supply Chain Management: An Interna- tional Journal 1(2): 28-35. Zakharchenko O. 2017. Ecological risks in agriculture: Problem-oriented approach. Business, Management and Accounting 3(1): 12-18.