Food Waste Management with Technological Platforms: Evidence from Indian Food Supply Chains

: Feeding the people sustainably continues to be a challenge in the present times. Enormous amounts of food wastage aggravate this problem. In developing countries, food wastage primarily occurs within the supply chain. Lack of technological infrastructure in these countries causes signiﬁcant post-harvest loss. While research shows that developments in food supply chains can reduce food wastage, no systematic research has been done so far to show the possible relationship between the use of technology and food loss. This paper attempts to address this gap by studying the supply chains of di ﬀ erent food processing organizations in India to assess the role of technological platforms in reducing food wastage in supply chains. Using a qualitative inductive methodology, the author identiﬁed the technological platforms that can address food wastage. Then, using multiple case-study analysis, the supply chains of sample ﬁrms were evaluated. The author assessed the food loss in these supply chains through comparative analysis to draw conclusions about the e ﬀ ectiveness of selected technological platforms. This study provides managers in the food industry with insights to prevent food loss, as well as some policy implications for developing economies. Overall, this paper throws light on the issue of food wastage and the possible means for its prevention.


Introduction
The food and agricultural systems of the world have been feeding more people than before. However, although more food is being produced, the problems of hunger and nutrient deficiencies are prevalent [1]. Aggravating this problem is the fact that about fourteen percent of food produced globally is lost during the post-harvest production stage [1,2]. In other words, 1.4 Gt of food suitable for human consumption is wasted each year [3,4].
Although in developed countries, food wastage mainly occurs at the consumer end, in developing countries, food wastage primarily occurs within the supply chain [5]. As this paper focuses on the ways to reduce food wastage in supply chains, the author limits the discussion to the food wastage in developing economies where the lack of infrastructure is the key reason for significant post-harvest loss [2,6]. Studies shows that post-harvest to distribution loss is highest in central and southern Asia, at nearly 21% [1,7]. Further, 85% to 90% of the observation points in central and southern Asia are from India, suggesting that food loss in supply chains are a major problem in the country. Indeed, several sources state that nearly 40% of the food produced in India is wasted [8][9][10][11].
A range of factors, such as microbial, enzymatic, chemical, physical, and mechanical ones, lead to food spoilage [12,13]. These factors necessitate the development of logistics systems in food supply chains [14]. Computerization and technological platforms facilitating online communications within food supply chains can facilitate the management of agricultural resources [15]. Researchers suggest that supply chains with advanced technological platforms can prevent nearly 50% of such loss [16].
(i.e., two professors of supply chain management, and one lecturer in food technology and engineering). Based on these insights and from prior literature, the author identified specific technological platforms that can improve the efficiencies in food supply chains: i.
Internet-based data monitoring and communication ii. Enterprise Resource Planning (ERP), i.e., software that helps integrate components of a company, including supply chain, by sharing and organizing information among participants at different levels [23] iii. Supply Chain Event Management (SCEM): this term refers to methods that process supply chain events [25]. In other words, it is a process of monitoring the planned sequence of activities systems along a supply chain and reporting any errors with the help of computerized monitoring devices [26] iv. Radio Frequency Identification (RFID) systems, i.e., small electronic tags that track the position and movement of items [27,28] v. Electronic Data Interchange (EDI), i.e., computer-to-computer exchange of documents for order processing, transactions, accounting, production, and distribution [23] vi. Programmable Logic Controller (PLC), i.e., a control system to monitor parameters of input devices and to generate decisions-based output parameters [23] vii. Cloud computing, i.e., an internet-based system to access a shared pool of computing resources (Mell & Grance, 2011) viii. Machine-to-machine, i.e., M2M communication or wireless or wired technology that captures data from a remote location using sensors and connects to the back-end enterprise systems via WLAN, satellite, or cellular communication [29,30].
In addition to the aforementioned technological platforms, literature described several other terms referring to the application of these platforms, such as logistics execution systems, network design applications, warehouse and transportation planning systems, and dashboard analytics for display and monitoring. This study includes these applications as well in order to have a holistic understanding of food supply chains and their technological infrastructure.
The conceptual framework for this study is described in the flowchart below ( Figure 1): This figure describes the considerations and steps leading to the data collection.

Materials and Methods
This paper uses qualitative analysis because qualitative study lends itself well to answer open ended questions such as the one studied in this paper [31]. Specifically, this paper uses qualitative inductive methodology [32,33] described by Gioia, Corley, and Hamilton (2013) [34]. This method captures the concepts relevant to organizational processes in an inductive manner, using grounded theory [20] building on procedures for open-ended inductive theory building [35]. In the early stage of the analysis, the author identified a number of informant terms, codes, and categories. Moving This figure describes the considerations and steps leading to the data collection.

Materials and Methods
This paper uses qualitative analysis because qualitative study lends itself well to answer open ended questions such as the one studied in this paper [31]. Specifically, this paper uses qualitative inductive methodology [32,33] described by Gioia, Corley, and Hamilton (2013) [34]. This method Sustainability 2020, 12, 8162 4 of 25 captures the concepts relevant to organizational processes in an inductive manner, using grounded theory [20] building on procedures for open-ended inductive theory building [35]. In the early stage of the analysis, the author identified a number of informant terms, codes, and categories. Moving forward, in the second order analysis, the author sought similarities and differences among the aforementioned categories. Wherever the author found concepts that were repeatedly present [36], these concepts were put together in the same themes. Finally, from these themes, broader dimensions related to the causes related to food wastage in supply chains (e.g., perishability, supply chain complexity) were identified. The author iterated between the data and literature several times, as suggested in the literature [36], in order to identify the relevant factors. Additional data was collected until a point of "theoretical saturation" was attained [20]. Prior literature was then used to explain these concepts and to use them further for analysis. This inductive analysis was followed by comparative analysis of the firms. Relevant parameters were rated in a comparative manner. Thereafter, the author formulated propositions based on the prior literature, and assessed the propositions on the firms. These steps are elaborated in the subsequent sections.

Sample Overview
In developing economies, food wastage primarily occurs in supply chains [5]. More specifically, post-harvest to distribution loss is highest in central and southern Asia [7], and significant in India [8,11], as described earlier. Hence, food supply chains in India were selected for this this study. To create a representative sample, the author chose organizations from different segments within the food industry and various parts of India. Their supply chains vary in complexity, volume, nature of the products, and market demands. Thus, the organizations are selected in a manner that takes their diversity and representativeness into account. For each firm, the author interviewed an employee who was closely involved with the products' supply chain. The interview questions are described in Appendix A. Table 1 below describes the studied sample. All company names have been disguised to maintain anonymity.
The Figure 2 below describes the location of each of the sample firms and shows their geographical spread.

Senior manager Delhi
The Figure 2 below describes the location of each of the sample firms and shows their geographical spread. Although the figure shows that a large number of sample firms are present in western India, it is an indicator of the representativeness of the sample. This is because western parts of India are more industrialized than the eastern region [39,40]. Consequently, there are more food processing firms and supply chains in western India as compared to the east.
After creating the sample of firms, the author proceeded with the data collection.

Data
The author took semi-structured interviews based on a questionnaire, allowing individual responses to guide further questions. The questions pertained to the following areas: the product and  Indian nutraceuticals (nutraceutical is a substance that may be considered food and provides medical or health benefits, including the prevention and treatment of disease [37]) for dietary supplements (a dietary supplement is a product containing a "dietary ingredient" intended to add further nutritional value to (supplement) the diet. A "dietary ingredient" may be one or a combination of substances [38] Although the figure shows that a large number of sample firms are present in western India, it is an indicator of the representativeness of the sample. This is because western parts of India are more industrialized than the eastern region [39,40]. Consequently, there are more food processing firms and supply chains in western India as compared to the east.
After creating the sample of firms, the author proceeded with the data collection.

Data
The author took semi-structured interviews based on a questionnaire, allowing individual responses to guide further questions. The questions pertained to the following areas: the product and its perishability, technological infrastructure in the supply chain, and food wastage. All interviews were in English. The author recorded each interview with permission from the respondent. Each interview lasted between 45 min and 1.5 h, with 75 min being the average interview duration. In total, 21.5 h of interview data was collected. The entire process of data collection lasted four months. Thereafter, the interviews were transcribed and used for analysis. Wherever required, the author collected additional information via additional discussions and email communication.
The author supplemented the primary data comprising of interview notes with secondary archival information available from the website of each company. This was followed by case-study writing for each firm. Wherever required, additional information from the interviewees was sought over email.

Analysis
The interview data suggested that food loss can occur at several stages in the supply chain, including raw material procurement, storage, production, dispatch, logistics, and retail. Moreover, different products have different requirements for processing and storage that may demand different forms of technologies. Based on preliminary interviews and prior literature, the author identified factors that determine the need for technological platforms in the supply chain to prevent food loss [12,41]. These factors are: i.
Supply chain complexity: Milgate (2001) [41] describes supply chain complexity as the uncertainty, technological intricacy, and organizational systems required to manage it. In other words, supply chain complexity refers to the number of production processes and needs for stringent control of processing and storage conditions [12,15]. From the interview responses, the author evaluated how complex the supply chains of each organization was. The author rated the sample firms on these parameters on three levels. For example, if the uncertainty in the processes (e.g., seasonality, reliance on weather), technological intricacy (need for advanced technology), and requirement of organizational systems were high (e.g., manual monitoring and supervision), the author classified supply chains as highly complex and rated as "5". This took into account the number of production processes and need for stringent control for processing and storage conditions. Similarly, if the processes did not have a high level of uncertainty, technological intricacy, and need for organizational systems, the supply chain was classified as "moderately complex" and rated as "3". Finally, if the processes had very low uncertainty, technological intricacy, and need for organizational systems, the supply chains were classified as "less complex" and rated as "1". Supply chains falling between these three levels were rated at 4 and 2, respectively. These ratings allowed us to evaluate the supply chains in a comparative manner and study the differences. ii. Perishability of raw material and product. According to the US Department of Agriculture [42], there is a likelihood for food to spoil, decay, or become unsafe for consumption if not maintained at specific conditions. Rahman (2005) and Singh and Heldman (2001) describe the criteria for the perishability of food products [12,43]. Based on these explanations, and based on the responses about the perishability of the raw material and final product as described by our respondents, they gave the following ratings: 1.

5.
Non-perishable: shelf-life of 12 months, rated 1 (Rahman, 2005). As understood from literature and interviews, these two parameters, i.e., supply chain complexity and perishability, largely determine the levels of food wastage when technological deployment is not taken into account. For example, high supply chain complexity can cause higher chances of operational errors that can lead to process failures and food loss. Additionally, high perishability makes raw materials or products vulnerable to spoilage easily when storage and processing conditions deviate from the required conditions. These two factors, i.e., supply chain complexity and perishability, were combined into another parameter, production sensitivity (PS). The author calculated production sensitivity as the sum total of "supply chain complexity" and "production sensitivity". Thus, the production sensitivity of the supply chains was rated on a scale of two to 10. Further, to analyze the firms in a comparative manner, the author classified them as either (i) having high production sensitivity if the production sensitivity rating is between 2 and 6, or (ii) having low production sensitivity, if the production sensitivity rating is between 7 and 10. This was followed by an assessment of the levels of food wastage.
All interviewees described food wastage in the supply chain as a percentage of production volume. The values of food wastage across all the sample firms ranged from~0% to 10%. Hence, food wastage levels were classified as follows: i.
The author also rated the firms according to the technology deployed at each organization. The author asked the interviewees whether the technological platforms described in Section 1.2 were being used at the organization or not. Table 2 presents the responses from each firm, and the corresponding rating. The same labels from Section 1.2 are used to refer to the technological platforms. Y denotes the presence of the technological platform while N denotes its absence. O denotes the presence of the technology at other locations. The presence of every technology contributed to 1 point in the rating, and the presence in another location contributed to 0.5 as it showed that the supply chain had technology, although it may not be entirely useful for the supply chain. The author gauged this qualitatively from the interview data. TR refers to the technology rating.  Finally, using production sensitivity (PS), the author evaluated the food wastage levels vis-à-vis the technology deployed at these organizations. From prior literature [12,43], the author found that when PS is high, supply chains require high technology deployment to prevent food wastage. If the production sensitivity is low, food wastage may be moderate or low even in the absence of technological deployment. These broad level understandings were tested on the data through the propositions presented in Table 3 below. In other words, Proposition 1 implies that when the production sensitivity is high, and technology deployment is high, prior literature predicts medium to low food wastage. Similarly, proposition 2 predicts that when the production sensitivity is high and technology deployment is low, there might be high food wastage in the supply chain. Propositions 3 and 4 can be interpreted in a similar fashion. These propositions were tested on the data gathered from the interviews and the prepared case-studies.

Results
The actual food wastage in each firm's supply chain was compared with the expected food wastage predicted from the propositions. Table 4 below describes the result.
Based on a comparative analysis of the expected food loss (according to the afore-mentioned propositions) to the actual food loss incurred by the firms in the samples, the following results were found: i.
Out of the 17 firms studied in the sample, 13 firms showed similar actual food loss as predicted in the propositions. A majority of the firms support the predicted propositions. Thus, the overall findings suggest that platforms can help prevent food wastage in supply chains. ii. Four companies, namely, C3, C10, C13, and C17, did not have actual loss as predicted by the propositions. The author studied them in further detail to understand the conditions in which technological platforms do not help in preventing food wastage, or are not required to avoid food loss. The following are the conclusions based on these four anomalous firms where the technological deployment did not show any relationship with the waste levels. iii. At C3, food loss levels are higher than expected despite having a high technological infrastructure in their supply chain. The respondent shared that the firm had not efficiently installed technologies, and there were "gaps [they] needed to fill". C3 was planning to initiate "process intensification" to explore how they could use "technological platforms to reduce food loss." The interviewee shared that they were "planning to use data from ERP to identify the areas of supply chain having higher wastage" and then "improve them for future". iv. At C10, actual food loss levels are lower than expected. The interviewee shared that "the highest chances of food loss are in the raw material storage, which requires temperature and humidity controls". The staff of C10 is trained to monitor the temperature and humidity regularly. "At the raw material warehouse, the likelihood of wastage is high." Hence, "the air-handling units [there] are automated". C10 also implements lean manufacturing, which ensures low wastage levels despite limited use of technology. v.
At C13, the waste is mainly the residue left after the flavors have been extracted from spices. However, C13 has been able to convert this "waste" into a by-product. They found alternate use of the residue as a "filler" in spice-blends. C13 sells the residue to spice mix companies. Hence, their wastage levels are very low, despite having a reasonably low technological infrastructure and high production sensitivity. vi. At C17, the food wastage levels are very low despite having limited technology and moderate production sensitivity. This low wastage level is because the company relies on manual observation and checks. However, the interviewee shared that "in future, implementation of technological platforms can reduce manual work, improve efficiency, and reduce errors".

Discussion
The overall conclusion from this study is that technological platforms can play a role in reducing food wastage in supply chains. For most firms, supply chain complexity and perishability of the raw material and products can serve as useful indicators to identify the relevance of technology.
Further, this study reveals that technological platforms can help reduce food wastage in supply chains, both directly and indirectly. The following examples show the direct effect of technological and other technology platforms in food supply chains.
First, automated PLCs reduce the chances of manual error and process failures. They also enable firm-level monitoring of a range of process parameters. Second, ERP helps identify efficient routing systems to improve logistics networks. Third, inefficiencies in the procurement system can be resolved with extensive backward integration. Technological platforms discussed in this study can facilitate the monitoring and control of such integrated supply chains to reduce wastage further. Finally, technologies like M2M communication enable significantly better control of ambient conditions. All these effects of technological platforms help in reducing food wastage in supply chains by enhancing operational visibility and process control. Moreover, this study also revealed several other indirect effects by which technological platforms can help reduce food wastage. For example, combinations of technologies like ERP and barcode readers enable the development of methods like "ready-make-discard". With such methods, retailers can identify and sell the earliest manufactured product unit. Such methods are crucial for supply chains like C8, where the products are highly perishable. This finding suggests that the scope of technological platforms in reducing food loss goes beyond the improvement of parameters like visibility, precision, and efficiency. Like in the case of C8, these systems can enable newer practices in supply chains to reduce loss of perishables. Also, technological platforms can help identify areas of high wastage. Thereby, firms can initiate efforts for process improvement, like in C3. Further, technological platforms can improve demand forecasting by connecting food manufacturers to retail stores or restaurants. For example, rapid demand fluctuations were a significant challenge at C4, which caters to the fast-food restaurant industry. Their product has low shelf life even under refrigerated conditions. With the help of internet-based technological platforms, C4 could communicate with their customers more efficiently. This implementation significantly reduced the food wastage in their supply chain. Finally, technological platforms can enable automation of certain processes. Although these processes may still require manual monitoring, they can reduce food wastage, as was noted in the C5.
Thus, technologies enable other processes that can indirectly reduce food wastage in supply chains. In sum, the study supports the proposition that technological platforms have the potential to influence food supply chains in a manner that would reduce wastage.
The findings from the present study are derived from qualitative data from one country. Although the author explained why India is an appropriate context for this study, it is worth acknowledging that the infrastructural challenges may be different in other parts of the world. Hence, future scholars can study the relationship between food wastage and technological platforms in other parts of the world such as South-East Asia and Sub-Saharan Africa where food wastage in supply chains is significantly high [2]. The author tried to make the sample as representative as possible, by incorporating firms of different industry segments, sizes, and location, there are limitations in the generalizability of the study owing to the qualitative nature of the data and analysis. Future research can study the relationship between technological platforms and the food wastage in supply chains quantitatively (for example, using linear regression and other related models). Further analysis can also reveal other the conditions under which technological platforms can be more or less feasible to firms, and the conditions under which they can be effective in reducing or preventing food wastage. Thus, this paper has opened up several avenues for future research.

Policy Implications
Reduced food wastage can potentially help in improving food security, reducing hunger, and malnutrition that are the critical issues in India and other developing economies. One purpose of this study was to generate evidence to support the potential of technological platforms in food wastage management in developing countries. Collaborative and synergistic actions are from the government and private sector to implement policies that can address the high volumes of food wastage in supply chains. Improving supply chains in the processed food industry can make a substantial difference in developing economies, as most wastage occurs after harvest, but before the produce or end-product reaches the final consumer. This study identifies a range of technological platforms that not only appear promising to address this problem but have proved their effectiveness in several food-processing companies in India. The investments made for the installation of such technologies can potentially be amortized with the savings from prevented food wastage in the long-run. The findings from this study can be applied to other developing economies that suffer from high food wastage and poor technological infrastructure in supply chains to address hunger and food security concerns across the world.

Conflicts of Interest:
The author declare no conflict of interest.

Appendix A. Interview Questionnaire
Introduction before the interview Dear Sir/Ma'am, My name is XXXX and I am doing research on food wastage in supply chains. With my research, I hope to better understand the gap in Indian industries leading to wastage and understand the infrastructure set up at companies of different operational scales. I will be very grateful if you could spare some time and answer this interview questionnaire. Please be assured that this thesis is for academic purpose and the information I gather will not be shared with any company. All individual and firm identities will remain confidential. Your response will be very valuable for my research. Thank you very much. What are the products manufactured by your firm? 2.
What is the scale of production in your company? Please specify whether this scale is for one facility and if there are multiple facilities owned by your firm.

3.
Please describe the SC of your main products in brief. (The remaining questions can be answered specifically for one or two major products produced by your firm) 4.
What is its shelf-life? 5.
Are there any special requirements of your food product? (In terms of pressure/temperature/ humidity or storage) a. At the time of production b.
At the time of transportation 6.
How are these ensured during the processing stage/transportation? 7.
How often is there a failure? (as an approximate estimate from your experience) 8.
In case of a failure, how are the personnel notified? How is the corrective action then taken? 9.
At what stage(s) of the production process/transport as the product susceptible to contamination? 10. What the ways by which any contamination can be detected? How is the personnel notified and corrective action taken? The case study of C1 is based on information from both primary and secondary sources (company website and social media pages). This company is the market leader in India in the several FMCG sub-sectors with a valuation of USD 9 billion and annual turn-over of over USD 3 billion. C1 Ltd. has set up a model called "e-Choupal" through which they have been able to integrate the primary producers (farmers in remote villages) with their supply chains by means of information technology (IT), thereby revolutionizing agriculture in places they have penetrated. ( Mandi refers to a marketplace in villages and towns where the farmers sell their produce at a price decided on the supply, demand, and seasonal patterns. Choupal: is derived from the Hindi word "Choupal", which refers to a traditional gathering place in villages. As in the case of communications technology, the prefix "e-" refers to electronic means of communication. Procurement: C1 has largely overcome the above-mentioned problems with the Choupal model set up in 1999. This model combines a web-portal in the local language (India has over 21 recognized languages and other regional languages) and personal computers with Internet access placed in the villages. This creates a communication channel between C1 and villagers. The produce is also collected from the farmers here. Information on prices of crops at different mandis, tips on farming practices, and so forth are also provided. This model has triggered higher payment based on produce quality and higher productivity. Choupal services today reach over 3. Total pan-India production:~85,000 TPM. Production and Distribution: Procurement through Choupal is described already. Further supply chain is described for a single product (Atta), in order to have brevity in the interview and case study. However, findings of other areas, such as IT implementation and food loss, are applicable to all food products. Quality checks for the raw material (RM) are done followed by grading, blending, and cleaning. Water-treatment softens the grains before grinding, which is done in three stages. This is checked by sieves, metal detectors, and microbial tests. Once packed, it is sent to warehouses, then to distributors, and then retailers. Key features of the RM: Wheat and other grains like soya bean, gram lentil, oat, and maize; sodium bicarbonate, ammonium bicarbonate, sugar, salt; vegetables, oils, lentils, and spices are the major raw materials. These have a shelf-life of 6-8 months. Relative humidity (RH) <60% is needed. Vegetables need refrigeration (used in 2-3 days). Key features of the product: Shelf-life varies from 3 to 12 months. All products are shelf-stable at ordinary temperatures. RH must not exceed 70%. Shelf-stability implies that the product does not lose its quality, does not degrade microbiologically or otherwise, and does not require refrigeration until after opening (USDA, 2014). Requirements of the production process and transportation: Temperature controls are required for almost every product during production. Failure occurrences and chances of contamination which could lead to food loss: Process fluctuation beyond control levels occurs once in 2 months or less. Features in the supply chain/production process to prevent/correct failure: RM is checked by QA department before usage. Sieves and metal detectors check the presence of contaminants. Temperature controls are automated using Programmable Logic Controller (PLC) set manually before every shift. If the failure still occurs, a signal notifies the personnel. WIP and FG are checked by QA. Food loss: Wastage is tracked by the systems used to track production and distribution (SAP for production, Astra and Sify for sales). Production loss is~0.7% of the production volume. Distribution loss account to~2% of the production volume. Technological infrastructure: E-choupal is used for procurement activities, which have their own server. SAP is used for production and logistics. For sales records and planning, "Sify" records data from distributor to retailer and "Astra" records date till the distributor. Dashboard Analytics is used in the Sales Process. GPS tracking is used for high value products like RTE meals, cigarettes, and personal care products. RFID is used till distributor warehouse and its data is fed directly into Sify and Astra. SAP was installed to coordinate the production with sales all over India, which is crucial, as there are over 1000 SKUs in food business alone. Due to SAP, supply chain efficiency and visibility have improved. Costs are optimized. Correlation between Food Loss and IT platforms: Prevention of wastage was one of the reasons for installing IT systems. Food wastage in factory was~3-4% earlier, now reduced 0.7% with IT integration. Marketing and distribution wastage has also dropped significantly, and there have been fewer product recalls. A major drop is wastage has been in the back end supply chain (from farmer to factory) with Choupal. Thus, there is a very strong correlation in reduction of wastage and implementation of IT platforms.
B2. C2 case (Based on the interview with Quality Assurance Officer, Operations)