Analysis of Severity of Losses and Wastes in Taiwan’s Agri-Food Supply Chain Using Best–Worst Method and Multi-Criteria Decision-Making ^†

Abstract

Food loss and waste are critical challenges in Taiwan’s agri-food supply chain, deteriorating security and resource efficiency. By employing the best–worst method (BWM), a multi-criteria decision-making model was developed in this study to evaluate the severity of losses and wastes. Combining literature review results with expert survey analysis results, key loss points, and mitigation strategies were identified to enhance sustainability and efficiency in Taiwan’s agricultural food system. Among the seven stages of the agricultural food supply chain, supermarket waste (16.95%) was identified as the severest, followed by government policies (16.63%), restaurant waste (15.35%), processing loss (14.71%), production site loss (13.64%), household waste (11.93%), and logistics/storage/distribution loss (10.79%). In the subcategories of each supply chain stage, the eight severe issues were identified as “Inadequate planning and control of overall production and marketing policies” under government policies, “Adverse climate conditions” and “Imbalance in production and marketing” under production site loss, “Inaccurate market demand forecasting” and “Poor inventory management at supermarkets” under supermarket waste, and “Improper storage management of ingredients leading to spoilage” as well as “Inability to accurately forecast demand due to menu diversity” under restaurant waste. The least severe issues included “Poor production techniques” under production site loss. Other minor issues included “Inefficient use of ingredients due to poor cooking skills”, “Festive culture and traditional customs”, and “Suboptimal food labeling design”, all of which contributed to household waste. Based on these findings, we proposed recommendations to mitigate food loss and waste in Taiwan’s agricultural food supply chain from practical, policy, and academic perspectives. The results of this study serve as a reference for relevant organizations and stakeholders.

1. Introduction

Food is a necessity for human survival. However, with the rapid growth of the global population, one in nine people worldwide still suffers from hunger, translating to approximately 800 million individuals in a state of hunger [1]. To alleviate the pressure of population growth, food production must be increased by 60% to meet the global demand for food [2]. Paradoxically, global food loss and waste have not decreased, with an estimated 33–40% of food lost or wasted annually, which is projected to reach 2.1 billion tons, equivalent to USD 1.5 trillion, by 2030 [3,4].

The severity of food loss and waste goes beyond unmet human needs, as it also exacerbates issues such as water resource depletion and greenhouse gas (GHG) emissions. According to the Boston Consulting Group (BCG), by 2030, GHG emissions from the food system would account for one-quarter of global emissions, with 18% of these emissions stemming from food supply chain processes such as processing and retail [5,6].

Recognizing the severity of food loss and waste, the United Nations included these issues in two of the 17 Sustainable Development Goals (SDGs) adopted in 2015. Specifically, Goal 12, sub-target 12.3, highlights the objective to reduce food loss and waste: “By 2030, halve per capita global food waste at the retail and consumer levels and reduce food losses along production and supply chains, including post-harvest losses.”

Taiwan has faced recurring issues of imbalance in the production and marketing of agricultural food. This problem is further influenced by Taiwan’s dense population, unique cultural and geographical characteristics, and household purchasing and consumption habits. Taiwan offers diverse sales channels for agricultural food, such as convenience stores, retail shops, traditional markets, and supermarkets. The high density and convenience of these purchasing channels often lead to increased food waste [7]. Given this situation, in this study, we identified and summarized the potential causes of food loss and waste within the agricultural food supply chain. By elucidating the current causes and severity of food loss and waste in Taiwan’s agricultural food supply chain, we suggested improvement measures for reducing food loss and waste based on the findings of the preceding analysis.

2. Literature Review

According to the definitions of Kusumowardani [8] and Parfitt [9], “loss” in agricultural food refers to the reduction in quantity occurring from initial production up to, but not including, retail sales. On the other hand, “waste” pertains to the discarding or loss of agricultural food occurring from the point of retail sales through to final consumption by end-users.

Borens [1] analyzed data published by the Food and Agriculture Organization (FAO) and summarized that more than 2 billion tons of agricultural food were lost or wasted annually. “Loss” accounts for 16% of total agricultural production, predominantly occurring during harvesting (6%), post-harvest handling and storage (5%), and processing stages (5%). On the other hand, “waste” constituted 14% of total agricultural production, occurring primarily at the retail (5%) and consumption (9%) stages. The interconnected nature of the agricultural food supply chain necessitates a systemic study to identify and address the underlying causes of food loss and waste across the entire chain.

The literature review results on agricultural food loss and waste, including works by Papargyropoulou [10], Yadav [11], Luo [12], Derqui [13], Filimonau [14], Sakaguchi [15], Tomaszewska [16], Ferro [17], Hebrok and Boks [18], Stancu and Lähteenmäki [19], Huang [20], Filimonau and Delysia [21], and Secondi [22] were used to classify the causes of agricultural food loss and waste into the following seven dimensions and related factors as follows.

• Production site loss: inadequate production techniques, extreme climatic anomalies, imbalances in production and marketing, and agreements with distributors;
• Processing loss: stringent standards for agricultural product procurement, improper inventory management of agricultural foods, inappropriate or erroneous processing methods, and suboptimal packaging design for processed agricultural foods;
• Logistics, storage, and distribution loss: inadequate logistics and storage/distribution infrastructure, improper logistics and storage/distribution management practices, insufficient capability of logistics personnel, and excessive food mileage;
• Supermarket waste encompasses waste: inaccurate forecasting of market demand, overly stringent inspection specifications, poor inventory management in supermarkets, and failed promotional activities;
• Restaurant waste: poor culinary skills leading to inefficient use of ingredients, an overly diverse menu making it difficult to accurately predict demand, improper storage management of ingredients resulting in spoilage, and exaggerated portion sizes aimed at attracting customers;
• Household waste is produced: inefficient cooking techniques leading to underutilization of ingredients, unplanned purchasing of food/ingredients, festive culture, and traditional customs, inadequate or malfunctioning household refrigerator storage conditions, and poorly designed food labeling;
• Government policies: overly stringent food safety regulations (e.g., shelf life, packaging standards, pesticide residue limits), inadequate planning and control of overall production and marketing policies, insufficient logistics-related infrastructure, and incomplete legal frameworks and supporting measures for the donation/sharing of near-expiry products.

This classification provides a comprehensive framework for understanding and addressing food loss and waste across various stages of the agricultural food supply chain.

3. Research Method

The multiple criteria decision-making (MCDM) method is a tool mainly used for decision-makers facing complex situations. It enables the formulation of decision analysis results aligned with the attributes and characteristics of the problem [23]. Among the various MCDM methods, the analytic hierarchy process (AHP) is the most widely used. The best–worst method (BWM) is conceptually similar to AHP but simplifies the pairwise comparison process, making it easier for decision-makers to evaluate and compare criteria [24]. The following steps outline the application of BWM in this study.

• Identify causes of agricultural food loss and waste: The main and sub-factors causing agricultural food loss and waste were determined and established as criteria for the BWM;
• Select the best and worst factors: Experts were invited to evaluate the causes of food loss and waste across various stages of the supply chain, selecting the most severe (best) and least severe (worst) factors at each stage;
• Compare other factors with the best factor: Experts rated the severity of other factors relative to the best factor on a scale from 1 to 9, where 1 indicated the least severe and 9 the worst;
• Compare other factors with the worst factor: The severity of the worst factor was rated as 1. The experts conducted pairwise comparisons to rate the severity of all other factors relative to the worst factor on a scale from 1 to 9;
• Calculate optimal weights: The optimal weights $(w_1^{\mathrm{*}},w_2^{\mathrm{*}},\dots ,w_n^{\mathrm{*}})$ for each factor were computed based on the BWM process;
• Consistency test: The consistency ratio (CR) was calculated to verify the internal consistency of each expert’s weight assessment, ensuring reliability in the evaluation process.

This structured method facilitated the prioritization of factors contributing to food loss and waste, providing actionable insights for addressing these issues systematically.

4. Results

19 experts were invited from relevant fields to complete the BWM questionnaire. The results derived from the BWM steps are summarized in

Table 1. Weighted factors contributing to food loss and waste in the agricultural food supply chain.

Main Factor	Weight (%)	Rank	Sub-Factor	Original Weight (%)	Adjusted Weight (%)	Rank
C1 Production site loss	13.64	5	C1-1 Poor production techniques	10.92	1.49	29
			C1-2 Extreme climatic anomalies	34.90	4.76	4
			C1-3 Imbalance in production and marketing	32.19	4.39	6
			C1-4 Distributor agreements	21.99	3.00	18
C2 Processing loss	14.71	4	C2-1 Stringent standards for product procurement	28.39	4.18	7
			C2-2 Improper inventory management	27.99	4.12	9
			C2-3 Inappropriate processing methods	24.05	3.54	13
			C2-4 Suboptimal packaging design	19.57	2.88	20
C3 Logistics, storage, and distribution loss	10.79	7	C3-1 Inadequate infrastructure	30.23	3.26	15
			C3-2 Poor management practices	28.07	3.03	17
			C3-3 Insufficient personnel capability	26.80	2.89	19
			C3-4 Excessive food mileage	14.90	1.61	28
C4 Supermarket waste	16.95	1	C4-1 Inaccurate market demand forecasting	34.14	5.78	2
			C4-2 Overly stringent inspection specifications	15.87	2.69	22
			C4-3 Poor inventory management	27.78	4.71	5
			C4-4 Failed promotional activities	22.21	3.76	12
C5 Restaurant waste	15.35	3	C5-1 Poor culinary skills	18.17	2.79	21
			C5-2 Overly diverse menu options	26.92	4.13	8
			C5-3 Improper storage management of ingredients	32.91	5.05	3
			C5-4 Exaggerated portion sizes	22.00	3.38	14
C6 Household waste	11.93	6	C6-1 Inefficient cooking techniques	16.08	1.92	27
			C6-2 Unplanned purchases	25.76	3.07	16
			C6-3 Cultural customs and celebrations	16.47	1.97	26
			C6-4 Inadequate or malfunctioning refrigerator	21.31	2.54	23
			C6-5 Poor food labeling design	20.38	2.43	25
C7 Government policies	16.63	2	C7-1 Overly stringent food safety regulations	15.20	2.53	24
			C7-2 Inadequate production and marketing control	38.84	6.46	1
			C7-3 Insufficient logistics infrastructure	22.84	3.80	11
			C7-4 Incomplete framework for near-expiry donations	23.13	3.85	10

This table consolidates the weighted importance of various factors contributing to food loss and waste ranked both at the primary and sub-factor levels.

.

5. Conclusions and Suggestions

Based on the BWM results, supermarket waste (16.95%) had the largest impact, followed by government policies (16.63%), restaurant waste (15.35%), processing loss (14.71%), production site loss (13.64%), household waste (11.93%), and logistics/storage/distribution loss (10.79%). The significant severity of supermarket waste in Taiwan was attributed to the high density of convenience stores, fresh produce supermarkets, and wholesale stores. With the increasing number of retail outlets, it becomes more challenging to accurately forecast sales at each location, exacerbating the problem of waste. Additionally, the BWM analysis results highlighted the critical role of government policies in mitigating loss and waste. This was evident not only in the sub-factor “C7-2 Inadequate production and marketing control policies” but also in “C1-3 Imbalance in production and marketing”, which is influenced by governmental planning. The severity of “supermarket waste” was largely caused by the intense retail competition and the difficulty in precisely predicting consumer demand in Taiwan’s dense retail landscape.

Based on the study results, the following recommendations were formulated from practical, policy, and academic perspectives to mitigate food loss and waste in Taiwan’s agricultural food supply chain.

1.

Practical perspective

As noted by Teng [7], the diversity of agricultural food sales channels in Taiwan contributes to the severity of supermarket waste. In Taiwan, small family businesses have become increasingly prevalent, which has shifted the primary sales channels to medium- and small-sized supermarkets and convenience stores. To address supermarket waste, a regionally integrated application (app) for near-expiry agricultural food products must be provided. This app allows supermarkets and convenience stores to consolidate and provide real-time information on near-expiry products available nearby, enabling consumers to make informed purchases and reducing waste at the retail level.

2.

Policy perspective

We identified “C7-2 Inadequate production and marketing control policies” as the sub-factor with the highest weighted importance. Although mechanisms such as production planning and price monitoring are implemented by Taiwan’s Council of Agriculture, their effectiveness remains limited. Effective government intervention, or the lack thereof, directly impacts the occurrence of loss and waste. It is therefore imperative for policymakers to refine and enhance these mechanisms, ensuring a more balanced and efficient alignment between agricultural production and market demand.

3.

Academic perspective

The definition of agricultural food encompasses a wide range of products, each with unique characteristics. Consequently, the post-harvest handling, processing, and logistical requirements vary significantly across different types of agricultural food. Future research is necessary to focus on specific agricultural products (e.g., rice) and leverage precise quantitative data to identify factors contributing to loss and waste. Such targeted studies enable actionable strategies tailored to the needs of specific agricultural products.