- 學位論文
食物廢棄物之永續管理最佳化分析—以臺灣為例
Sustainable Optimization Analysis of Food Waste Management： A Case Study in Taiwan
摘要
聯合國糧農組織於2011年的報告中指出，人類每年扔掉的食物約佔了糧食總產量的三分之一。食物廢棄物對於環境更是有著巨大且不容忽視的影響，食物廢棄物之管理也顯得格外的重要。而各地區食物供應鏈上各階段產生食物廢棄物的型態依照各地條件而有差異，因此在研究地區性之食物廢棄物管理時，需特別注意各地間食物供應鏈特性之差異。目前許多研究都以食物廢棄物分層等級作為出發點以評估食物廢棄物管理之決策，亦應用生命週期評估分析方法至此議題上，但大多聚焦在食物廢棄物下游處置上，用以評估及決策的方法仍是相當的有限。本研究以臺灣為主要研究地區，以生命週期評估分析臺灣食物供應鏈各階段之環境衝擊，並同時利用線性規劃法，結合環境因子及經濟因子，建立了兩組臺灣食物廢棄物永續管理最佳化模式，分析如何以預防、重複利用及回收再利用共三大類16項管理方案配置處理食物廢棄物以達到最大環境效益，以建立更符合在地化情況之食物廢棄物管理決策輔助工具。 結果顯示在臺灣，榖類、肉類及蔬果類食物皆以生產階段為衝擊熱點，乳類之衝擊熱點為加工階段，並且皆以氣候變遷之人體健康、化石燃料耗用及粒狀污染物形成為佔總衝擊中最多比例之三類衝擊類別。最終處置方式部分，總環境衝擊最大者為集中堆肥，次者為厭氧共消化，接著是動物飼料及焚化，其衝擊結果受是否考量能資源回收之替代效益影響甚大。於最佳化結果顯示，消費者教育活動、統一日期標示、產品規格改變、改善存貨管理、冷鏈管理、標準化捐贈規範及捐贈義務教育皆兼具環境效益及經濟有效性。而捐贈租稅誘因、標準化捐贈規範、捐贈運輸設施、加值處理及捐贈義務教育中應用於消費階段時所能產生之環境效益為負值，意即以這些方案處理消費階段產生之剩食的環境衝擊大於將這些剩食進行焚化處理之環境衝擊，較不具環境優勢，捐贈倉儲與管理以及集中堆肥處理則為在模擬結果中最不具環境效益之管理方案。
並列摘要
Food and Agriculture Organization of the United Nations (FAO) has estimated that about 30% of the food produced for human was lost or wasted along the food supply chain. Food waste is increasingly acknowledged as an urgent issue globally with not only a social cost but also negative consequences on the environment and natural resources. Therefore, comprehensive analysis of food waste managements is expected to make a significant contribution to achieve sustainable development goals. The characteristics of food waste along the food supply chain varies tremendously from place to place, taking the differences into account for the decision making of food waste managements is rather necessary. Life cycle assessment (LCA) is an applied approach for evaluating environment impacts over the whole life cycle of a product, process, and activities. LCA is widely used as an assessing method in supporting systems for quantifying the environmental impacts for waste managements. The waste hierarchy h prioritize waste management measures by their environmental preferences and has recently been applied to the management of food waste. Up till now, studies for food waste management has been focusing on assessing the disposal or recovery methods of the food waste produced in the consumption stage, regardless of the prevention methods, reusing methods, and the food waste produced in other stages in food supply chain. Also, there was limited knowledge on the evaluation of food waste management strategies including both economic and environmental aspects. This study aimed at establishing two linear programing models for the optimization of food waste management. By considering both environmental and economic indicators, the approach developed in this study identified the optimal combination of food waste management options in order to maximize the environmental benefits along the food supply chain in Taiwan within certain budget. LCA results showed that the hotspots for environmental impacts of food supply chains in Taiwan was production stage for grain, meat, fruits and vegetable, and process stage for milk. On the other hand, the optimization result showed that, among all options, donation storage and handling and centralized composting were the least favorable food waste management measures in Taiwan, which indicated that these two options would be excluded at the planning stage of the food waste management in the case study area. With more comprehensive consideration of the life cycle of each type of food waste, and also more localized life cycle inventory data, a more authentic LCA approach was established along with the optimization modeling established to identify the optimal combination of each treatment quantity, thereby creating a more effective decision support tool in food waste management. This study seeks to enhance the authenticity and reliably of existing food waste management analysis by improving the data inventory of LCA, the omitted impacts from upstream stages of food wastes and also the optimization modeling into actual amount treatment quantity, which can further be utilized by stakeholders to be more initiative faced with multiple management options.