Integrated approaches to circular bioeconomic systems considering supply chain sustainability and resilience in agriculture

Uen, Tinn-Shuan

Permalink

https://hdl.handle.net/2142/129771 Copy

Description

Title

Integrated approaches to circular bioeconomic systems considering supply chain sustainability and resilience in agriculture

Author(s)

Uen, Tinn-Shuan

Issue Date

2025-05-02

Director of Research (if dissertation) or Advisor (if thesis)

Rodríguez, Luis F.

Doctoral Committee Chair(s)

Rodríguez, Luis F.

Committee Member(s)

• Ting, Kuan-Chong
• Zhang, Yuanhui
• Ouyang, Yanfeng

Department of Study

Engineering Administration

Discipline

Agricultural & Biological Engr

Degree Granting Institution

University of Illinois Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Supply chain optimization
• Circular Bioeconomy
• Resilience
• Sustainability

Abstract

The circular bioeconomy has gained greater attention as a framework to address significant challenges of food waste and resource efficiency in food supply chains (FSCs). By reusing wasted resources and mitigating environmental impacts, this approach creates new opportunities to boost sustainability, particularly in light of substantial effects of food waste on climate change. In response to rising demand uncertainty and unexpected disruptions, especially for perishable products, resilience and sustainability have become essential to FSC operations. Although currently underdeveloped, systematic approaches for sustainable food waste management, such as generating renewable resources like bioenergy and biofertilizer, present promising opportunities to enhance bioeconomies within FSCs. Quantifying regional food waste generation and distribution under stochastic demand can expand recycling opportunities and improve the efficiency of food waste collection. Moreover, understanding trade-offs between profitability and resilience across varying FSC designs can inform network and logistics planning, facilitating robust strategies against disruptions. Therefore, this thesis aims to develop systematic and scalable approaches to strengthen circular bioeconomic resilience in FSCs. To enhance food waste valorization while reducing greenhouse gas emissions of a reverse FSC, from consumption to bioprocessing and reuse, an integrated approach has been proposed that combines geospatial analyses, a clustering algorithm, and mixed-integer linear programming. This model maximizes net present value by optimizing logistics and configuration for food waste management in Illinois, using stand-alone anaerobic digestion and co-digestion at wastewater treatment plants. The findings reveal an average return on investment of 8.3% with an annual reduction of 1,012 thousand Mg of CO2 equivalents. Co-digestion is shown to be more economically competitive than stand-alone digesters. Tipping price, capital investment, operational cost, and food waste availability are identified as the most significant factors influencing the net present value of the system. Predictive models have been developed to estimate county-level food waste generation using machine learning models. This study considered factors such as agricultural sales, consumption, demographics, grocery store density, income patterns, and access to social programs. The results show that the Support Vector Regressor achieved the best overall performance with an R-squared value of 0.837, median mean absolute error of 792 metric tons, and root mean square error of 3,385 metric tons. The developed models accurately forecast food waste generation for most US counties with levels at or below 60 thousand metric tons, though they tend to underestimate waste generation in large counties such as Cook County, IL, and Los Angeles County, CA. The most influential feature is the number of food stores and restaurants, followed by population, the number of stores with food support programs, and animal product sales. These outcomes will support resource collection and valorization planning across various waste streams, thereby supporting the development of circular bioeconomies. Beyond addressing food waste challenges in a reverse FSC, mathematical optimization models have been developed to evaluate economic and resilience performance of a forward FSC, from production to processing and markets, in the Midwestern beef supply chain. Centralized and decentralized networks were designed by maximizing total system profits under two capacity settings for meat processors: existing capacity for the centralized network and medium capacity for the decentralized network. Profitability and resilience in centralized and decentralized beef supply chains under various disruption severities and ripple effects. The findings indicate that a centralized network is more profitable under normal conditions, with an annual advantage of 1.42 billion USD over a decentralized network. However, decentralized networks consistently outperform centralized networks in profitability and beef product supply across disruption scenarios. Results from disruption impact analysis and post-disruption optimization reveal that decentralized networks offer greater resilience, requiring fewer resources to maintain profitability and product supply during disruptions. This study provides valuable decision support for supply chain stakeholders seeking higher resilience under unforeseen circumstances. As network designs in FSCs influence food waste distribution, this study bridges the gaps between food waste generation and FSC designs by proposing a stochastic rolling horizon model to optimize weekly inventory control for markets in the Midwestern beef supply chain while quantifying food waste under market demand uncertainties. The results indicate that fluctuations in market demand can generate food waste costs of approximately 300 million USD under an ±80% uncertainty range, with food waste costs exhibiting a non-linear relationship with demand uncertainty. Cost breakdown analysis reveals that unmet demand incurs higher costs than food waste, but food waste is more sensitive to market demand fluctuations. Variations in inventory and food waste costs peak at a ±50% demand fluctuation level. This study provides a useful benchmark inventory model to quantify food waste at markets for perishable food products, facilitating more accurate estimates of retail and market-level food waste.

Graduation Semester

2025-05

Type of Resource

Thesis

Handle URL

https://hdl.handle.net/2142/129771

Copyright and License Information

Copyright 2025 Tinn-Shuan Uen

Owning Collections