Progress 06/15/24 to 06/14/25
Outputs
Target Audience: The audiences include (i) The staff, volunteers, and clients of Alabama food pantries, with most clients experiencing food insecurity and coming primarily from low-income and low-food access communities. (ii) Agricultural and Engineering students at Auburn University (AU) and Tuskegee Universities (TU); (iii) Alabama farmers who have surplus food on their farms and are interested in the Farm to Food Bank program (iv) People/stakeholders who care about the food and nutrition insecurity issues, including food banks, Feeding America, USDA, food panties, etc. Our Efforts include (i) Developing a new machine learning course, that provides experiential learning opportunities for Tuskegee University (TU) and Auburn University (AU) students by integrating real-world research projects into the curriculum and incorporating field visits. (ii) Designing machine learning algorithms for food demand forecasting and food inventory detection; (iii) Designingsurveys to assess food insecurity; (iv) Creating a GIS map, webpages, and an APP for information dissemination; (v) Organizing training programs and events for nutrition intervention; (vi) Developing a pilot program for the farm-to-food bank/pantry. Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided?Training Activities: The project has supported the training of one Ph.D. student at Auburn University and one Ph.D. student at Tuskegee University for their dissertations/essays. This project trained approximately 29 students from Auburn and Tuskegee Universities through the proposed inter-campus course, Applied Statistics and Machine Learning. The project supported three undergraduate students from Auburn University to work on image labeling and coding for food image detection under Goal 2. The project supported onePh.D. student from Tuskegee University to collect food image data from a campus food pantry. This experience helped train the student in communication with food pantry staff and volunteers, as well as in food image collectionskills. Professional Development: The project supported two Ph.D. students in attending and presentingat the 133rd Annual Farmers Conference in Montgomery, AL. The project supported onePh.D. studentin attending and presentingatthe 82nd Annual Professional Agricultural Workers ConferenceinMontgomery, AL. The project supported five students from Auburn and Tuskegee Universities to attend the annual Goat Day event, where they learned how AI and machine learning can help solve real-world issues related to goats and goat meat. How have the results been disseminated to communities of interest?We disseminated our research findings to food banks and researchers through presentations at the 2024 Professional Agricultural Workers Conference (PAWC) and the 2025 Farmers Conference. The presentations attracted more than 60 attendees, including farmers, government officials, USDA officers, professionals, and others. They helped raise awareness of the need for AI and digital transformation for food banks and food pantries and demonstrated how machine learning algorithms can improve the operational efficiency of community food assistance programs, ultimately enhancing food and nutrition security. We maintain close communication with the AUMC and Lakeview food pantries, particularly with Mr. Joe Davis, and Ms. Megan Coppenger. Their support has been essential in accessing the latest datasets, learning operational practices, and confirming local policies and implementation details referenced in our paper. Our econometrics and machine learning-based food demand forecasting, along with the computer vision-based food image detection algorithm, have been customized to meet the specific needs of their operations. Ultimately, we hope these decision-support AI and machine learning methods will be adopted in their operations to improve overall efficiency. What do you plan to do during the next reporting period to accomplish the goals?For Goal 1: During the next reporting period, we plan to provide food demand forecasting services to local food pantries. Specifically, we will develop a user interface (UI) that allows pantry staff to input updated data and receive on-demand forecasts based on our forecasting models. We also plan to explore the scalability of this service to other food pantries across Alabama. Additionally, we will map the food pantry locations and make them publicly viewable to ensure accessibility for those in need. For Goal 2: During the next reporting period, we plan to train and evaluate the YOLOv11 model using our annotated food pantry dataset. In parallel, we will explore Florence-2 for zero-shot food item detection to assess its generalization capability without additional training. We also aim to link detected food items to nutrition databases and prepare a journal paper detailing our findings and applications for food inventory and nutrition analysis of distributed food from food pantries. For Goal 3: We plan to collaborate with the Tuskegee University Cooperative Extension Program (TUCEP) and provide one or two trucks to help transport surplus edible food from farms to food banks after harvest. This effort will assist in transferring surplus nutritious food to individuals in need and enhance the capacity of Alabama's Emergency Food Distribution System to provide healthy food to food-insecure populations. Meanwhile, farmers will receive tax benefits from their donations, as well as goodwill and strengthened relationships within their local communities. For Goal 4: We will continue to offer and enhance the Applied Statistics and Machine Learning course, while expanding experiential learning opportunities for students at Tuskegee University (TU) and Auburn University (AU). For example, we will introduce more real-world machine learning projects into the course as case studies or course projects, and maintain field visits as hands-on experiences for the students. Additionally, we plan to establish a website for this project to share our findings with the public, raise awareness about the importance of community food assistance programs (such as food banks and food pantries), and demonstrate how our project can help these organizations better serve people in need.
Impacts
What was accomplished under these goals?
1) Goal 1 - Machine Learning-Based Food Demand Forecasting at Alabama Food Pantries Under this goal, we plan to submit a paper titled "Data-Driven Food Demand Forecasting at Food Pantries in Alabama" to the peer-reviewed journal in June. This study employs three traditional time series models and eight machine learning (ML) algorithms to forecast next-week food demand--measured by household visits and distributed food weight--using data from 2016 to 2024 collected at two Alabama food pantries. To address challenges such as small sample sizes, high weekly variability of food demand, and external shocks, we incorporate contextual factors, including pantry closures, calendar features, and socioeconomic indicators. Results show that simple linear ML models consistently outperform both current operational baselines used by the pantries and other alternative models across multiple forecasting tasks. Furthermore, SHAP (SHapley Additive exPlanations) analysis identifies pantry closures, SNAP participation, and calendar-based features as key drivers of food demand. Incorporating closure indicators and cyclical temporal encodings also significantly improves forecasting accuracy. Overall, this study offers scalable, data-driven forecasting tools that can be adopted by other food pantries in Alabama to enhance operational planning, reduce food shortage and surplus, and improve the timeliness and targeting of food assistance programs. Two metrics are used to evaluate the models's forecasting performances: Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE). At the Lakeview Food Pantry, the Ridge model, incorporating current and lagged values of the food demand variable along with social, demographic, economic, and COVID-related variables, achieved the best forecasting performance for household visits (MAE: 6.9; RMSE: 8.8), and distributed food weight (MAE: 366.2; RMSE: 460.3). Compared to the existing operational method used by the pantry, our approach achieves a 43% reduction in MAE and RMSE. At the AUMC Food Pantry, the Gradient Boost model using the same control variables also performed best in forecasting household visits (MAE: 8.7; RMSE: 11.6), and distributed food weight (MAE: 419.5; RMSE: 520.9). These results represent a 28% reduction in MAE and a 22% reduction in RMSE compared to the pantry's current operational forecasting method, highlighting the potential of machine learning tools to enhance efficiency and decision-making in community food assistance programs. 2) Goal 2 - Develop machine learning and digital technologies to detect food supply in Alabama Emergency Food Distribution Centers toward a higher operational efficiency environment. Under this goal, we focused on preparing a robust dataset to support the future development of an automated food item detection system. We collected and annotated 1,123 images containing 13,795 food items across four food pantries: Auburn United Methodist Church (AUMC) Food Pantry, Lakeview Food Pantry, Auburn University Food Pantry, and Tuskegee University Food Pantry. All food items were categorized into 25 food categories. We are currently planning to implement advanced object detection models such as YOLOv11 and Florence-2, with the goal of enabling automatic food item detection and future nutrition analysis to improve inventory tracking and food distribution efficiency. In addition, we completed and submitted a research paper titled "Explainable Deep Learning for Meat Freshness Detection" to the journal Artificial Intelligence in Agriculture (currently under review). The methods and insights developed in that study support the technical foundation for food classification and explainability in this goal. 3) Goal 3 - Increase the Alabama Emergency Food distribution system's capacity for providing healthy food to food-insecure populations. We co-hosted a session titled "From Farm to Food Bank: Empowering Rural Prosperity and Food Security" at the 133rd Annual Farmers Conference in Montgomery, Alabama. The session focused on reducing food waste through agricultural donations, improving food access for individuals in need, and fostering partnerships between producers and emergency feeding organizations. Two invited speakers gave presentations: Dr. Don Wambles, Director of the Ag Promotions/Farmers Market Authority Division at the Alabama Department of Agriculture and Industries, and Tracey Guidry, Grants Manager at the Heart of Alabama Food Bank. The two presentations raised awareness of community food assistance programs, specifically the "Farm to Food Bank" project, and informed farmers and other attendees about how farmers and others can benefit from donating surplus edible food to food banks. They also provided guidance on how to donate unsold, yet still edible, food supplies to local food banks. In addition, two students presented their preliminary research findings. One presentation focused on food demand forecasting, introducing methods to improve the accuracy of predicting food needs for local food pantries in the upcoming open week. These methods have the potential to enhance local pantries' ability to distribute healthy food to food-insecure populations more efficiently. The other presentation explored the use of advanced object detection models, such as YOLOv11 and Florence-2, to enable automatic food item recognition. This work lays the foundation for future nutrition analysis, improved inventory tracking, and more efficient food distribution. Moreover,we also submitted a research paper titled "Explainable Deep Learning for Meat Freshness Detection" to the journal Artificial Intelligence in Agriculture (currently under review). The study focuses on using AI to assess food quality and freshness, which directly aligns with the goal by contributing tools and insights that help ensure only high-quality, nutritious foods are distributed to vulnerable populations. 4)Goal 4 - Strengthen AI/Machine Learning Education at Tuskegee University PD Chen and Co-PD Sun developed and enhanced the new course titled"Applied Statistics and Machine Learning", which was offered to both Tuskegee and Auburn undergraduate and graduate students in Spring 2025. This is the first machine learning course offered by the College of Agriculture, Environment & Nutrition Sciences (CAENS) at Tuskegee University. It also serves as a core course for both the new inter-college Data Science degree program and the new microcredential in Applied Machine Learning, both of which were launched at Tuskegee in 2024. A total of 29 students were enrolled in the course that semester, including 20 from Auburn and 9 from Tuskegee. This USDA-NIFA project supported six-course projects within the class, helping to strengthen AI in Agriculture research and education at Tuskegee University. Each project included both Tuskegee agricultural students and Auburn engineering students to balance their expertise. They learned from one another and applied the machine learning methods they had learned from class to address real-world challenges faced by agricultural and rural communities.It also fostered stronger collaboration between Tuskegee, Auburn, and the local community.
Publications
- Type:
Other Journal Articles
Status:
Under Review
Year Published:
2025
Citation:
Bramantyo, H. A., Faridi, M. A., Chen, R., Harris, C., & Sun, Y. (2025). Explainable deep learning for meat freshness detection. Artificial Intelligence in Agriculture, under review.
- Type:
Peer Reviewed Journal Articles
Status:
Published
Year Published:
2024
Citation:
Chen, R., Y. Su, and L. Tran. 2024. Small Farmers Perceptions of Climate Change and Adoption of Climate-Smart Practices: Evidence from Missouri, USA. Sustainability 16(21):9525.
- Type:
Conference Papers and Presentations
Status:
Accepted
Year Published:
2024
Citation:
Mengxue Li, Rui Chen, Yin Sun. Common Factor Augmented Food Demand Forecasting for Alabama Food Pantries." The 82nd Annual Professional Agricultural Workers Conference (PAWC), November 2024, Montgomery, AL.
- Type:
Conference Papers and Presentations
Status:
Accepted
Year Published:
2025
Citation:
Mengxue Li, Rui Chen, Yin Sun. Demand Forecasting at Alabama Food Pantries Using Machine Learning Methods. The 133rd Annual Farmers Conference, March 2025, Montgomery, AL.
- Type:
Conference Papers and Presentations
Status:
Accepted
Year Published:
2025
Citation:
Hutama Arif Bramantyo, Rui Chen, Yin Sun. Food Image Detection: AI Methods, Nutrition Link, and Impact. The 133rd Annual Farmers Conference, March 2025, Montgomery, AL.
- Type:
Peer Reviewed Journal Articles
Status:
Published
Year Published:
2024
Citation:
Chen, R., D.T. Adu, W. Li, and N.L.W. Wilson. 2024. Virtual water trade: Does bilateral tariff matter? Ecological Economics 222:108216.
|
Progress 06/15/23 to 06/14/24
Outputs
Target Audience:The audiences include (i) The staff, volunteers, and clients of Alabama food pantries, most of the clients are from socially disadvantaged communities. (ii) Agricultural and Engineering students at Auburn University (AU) and Tuskegee Universities (TU); (iii) Alabama farmers in Alabama who have food surplus on their farmers; (iv) People/stakeholders who care about the food and nutrition insecurity issues, including food banks, Feeding America, etc. Our Efforts include (i) Developing a new machine learning course, providing experiential learning opportunities for TU and AU students. (ii) Designing machine learning algorithms for food demand forecasting and food inventory detection; (iii) Conducting surveys; (iv) Creating a GIS map, webpages, and an APP for information dissemination; (v) Organizing training programs and events for nutrition intervention; (vi) Developing a pilot program for the farm-to-food bank/pantry. Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided?Training activities:The project has trained two PhD students at Auburn University andone PhD student at Tuskegee University. In addition, the project provided two course projects for the "Applied Statistics and Machine Learning" course that PDChen and Co-PDSun offer during the Spring 2024 semester to undergraduate and graduate students at both Auburn University (AU) and Tuskegee University (TU). One project is "food demand forecasting at the Auburn United Methodist Church (AUMC) food pantry," and the other project is"computer vision-based food item classification." Seven students worked on these two course projects and presented their work through poster presentations.The students of the "Applied Statistics and Machine Learning" course visited the AUMC food pantry as a field visit for experiential learning. PDChen and Co-PDSun co-advised one REU studentfrom UIUC in the Summer of 2023.The UIUC undergraduate studentgave two poster presentations on "The Image Classification of Vegetables for Tracking Food Pantry Inventory," one at the Tuskegee University/UIUC REU/REEU seminar and the other at the AIFARMS annual meeting. Professional development: Support one Ph. D student to participate in Post Conference (2024 AAEA Annual Meeting) Workshop on Food Prices and Forecasting Sponsored by ERS (PC52) How have the results been disseminated to communities of interest?We disseminate our research ideas and results to the students ofour"Applied Statistics and Machine Learning" course. Our students from the course and the REU summer training program presented their work at Auburn University, Tuskegee University, and the AIFARMS annual meeting. We are partnering with the AUMC food pantry and have been communicating closelywith Mr. Joe Davis, theDirector of Mission and Outreach at AUMC about our research progress. We will reach out to more food pantries for data collection and get their feedbackas we prepare journal articles that reportour research outcome. What do you plan to do during the next reporting period to accomplish the goals? For Goal 1, "Food Demand Forecasting for Local Food Pantries," the preliminary results of several machine learning forecasting algorithms show the best performance at 25.67% error due to limited data points. We will continue to improve our forecasting results by augmenting our AR model with estimated common factors using the Principal component analysis (PCA)method. This approach will allow us to address high-dimensional time series data by reducing dimensionality and identifying latent factors that capture significant variations. We aim to publish this work. For Goal 2, "Food Inventory Detection at Local Food Pantries," we aim to prepare a research paper for publication. To achieve this, we will visit multiple food pantries across Alabama to gather a diverse dataset representing various food pantry environments and expand the number of food product classes. Meanwhile, we will employ a more advanced image classification model, such as YOLOV8, for the training and testing of our algorithm. For Goal 3, the PD and Co-PDs have been planning to collectfood boxes from these pantries and provide healthy cooking skillsfor affordable food items included in the boxes. We plan to host 1-2 workshops on food nutrition education for food pantry clients, staff, and the food-insecure population at local food pantries and/or Farmer Conferences. Additionally, we have reached out to local farmers and obtained their approval to donate leftover produce from farms to food pantries. We will establish a pilot program, "From Farm to Food Pantry," to assist in transferring fresh food produce from farm to food pantry. For Goal 4, we will continue to provide experiential learning opportunities for TU and AU students by collaborating with local food pantries for student visits. Additionally, we will integrate new machine learning algorithm materials from the projects in Goals 1 and 2 into our new course, 'Applied Statistics and Machine Learning,' to enhance students' knowledge and improve the outcomes of their course projects.
Impacts
What was accomplished under these goals?
1) Goal 1 Machine Learning-based food demand forecasting at Alabama food pantries: Under the first goal, which aims to successfully forecast household visits to food pantries, understand food demand at Alabama food pantries, and benefit the food pantries' operations, we developed seven machine learning algorithms. In our preliminary work, we constructed a dataset with household visits as the target variable and 17 independent socio-economic variables, such as CPI, unemployment rate, COVID-19 cases, school days, and more. We collected historical household visits data from two food pantries: AUMC and Lakeview. By communicating with the food pantry directors, we obtained detailedinformation about missing data, discrepancies, and recent updates to ensure the accuracy ofhousehold visits data. The seven machine learning algorithms used to forecast household visits are: (1) Linear Regression; (2) LASSO Regression; (3) Neural Network based Regression; (4) Gradient Boosting; (5) Decision Tree Regression; (6) Bayesian Ridge Regression; (7) Ridge Regression. The preliminary results and source codes have been shared with the public in an open-source GitHub repository: https://github.com/Kamran0153/Food. The online table shows the performance of machine learning algorithms measured by Mean Absolute Percentage Error (MAPE). Baseline algorithms have high inference errors of 71.4% and 57%. Gradient boosting, without Variational Auto-Encoder, achieves a 38.44% error. Using Variational Auto-Encoder reduces errors by 10%-20%, with Neural Network based Regression performing best at 25.67% error. We are currently exploring PCA methods to understand the impact of COVID-19 on our time-series data and improve the accuracy of food demand forecasting. We employ an augmented AR model combined with 6 latent common factors, which are derived using the PCA method. The results show that this model performs better than the benchmark model. 2) Goal 2 Machine Learning algorithms for food item detection at Alabama food pantries: Under the second goal, we have developed a machine learningalgorithm thatautomatically detects food items from images taken in pantry settings. Initially, we visited the Auburn University Methodist Church(AUMC) Food Pantry multiple times and collected our dataset. We collected a total of 220 images from AUMC Food Pantry, each image captures a collection of different food items placed randomly on a table. Wedivided thisdataset into three parts, i.e.,training (88%), validating (6%), and testing (6%) our model. UsingRoboflow's "Smart Polygon" tool we have labeled our dataset and identified 33 different classes of food products.For our machine learning model, we have utilized Roboflow's free Image Segmentation model, "Roboflow 3.0 Instance Segmentation (Fast)" which uses the "Segment Anything Model (SAM)" in the background. After testing, our machine learning model resulted in a precision of 77.1% and a Mean Average Percentage Error(MAPE)of 22.41%. We will collect more images for different food pantries to improve the accuracy of our algorithm. The outcomes under Goals 1)-2) can be used to improve the operations of Alabama food pantries. Specifically,accurate food demand forecasting allows food banks and food pantries to optimize inventory management, thereby enhancing the overall efficiency of the emergence food distribution system.The automatic food item detection model can be used for nutrition analysis and food inventory monitoring. We will study these directions next. 3) Goal 3 Increase the food pantry's capacity to provide healthy food: We have obtained approvals from local farmers to donate fresh food surplus from their farms to food pantries, as well as approval from the local food pantry to receive these donations. We will launch a pilot program, 'From Farm to Food Pantry,' in Year 2. 4) Goal 4 Strengthen the AI/machine learning education at Tuskegee University:PD Chen and Co-PD Sun developed a new "Applied Statistics and Machine Learning" course, which was offered to Tuskegee and Auburn undergraduate and graduate students in Spring 2024.ThisUSDA-NIFA project provided two-course projects for the "Applied Statistics and Machine Learning" course. One project is "food demand forecasting at theAuburn United Methodist Church (AUMC) food pantry," and the other project is"computer vision-based food item classification." Seven students worked on these two course projects and presented their work through poster presentations. The students of the"Applied Statistics and Machine Learning" course visited the AUMC food pantry as a field visitfor experiential learning.These activities canenhance "AI in Agriculture" research and education at Tuskegee University and foster stronger collaborations between Auburn, Tuskegee, and local community.
Publications
- Type:
Journal Articles
Status:
Published
Year Published:
2023
Citation:
Yan, Kevin, Md Kamran Chowdhury Shisher, and Yin Sun. 2023. "A Transfer Learning-Based Deep Convolutional Neural Network for Detection of Fusarium Wilt in Banana Crops." AgriEngineering 5, no. 4: 2381-2394. https://doi.org/10.3390/agriengineering5040015.
- Type:
Journal Articles
Status:
Published
Year Published:
2024
Citation:
Chen, Rui, Derick T. Adu, Wenying Li, and Norbert L.W. Wilson. 2024. "Virtual Water Trade: Does Bilateral Tariff Matter?" Ecological Economics 222: 108216. Elsevier.
- Type:
Conference Papers and Presentations
Status:
Accepted
Year Published:
2023
Citation:
Chen, Rui, Md Kamran Chowdhury Shisher, Thomas Orrison, and Yin Sun. "Demand Forecasting at Alabama Food Pantries Using Machine Learning Methods." Southern Economics Association 93rd Annual Meeting, 178. November,2023. New Orleans, LA.
- Type:
Conference Papers and Presentations
Status:
Accepted
Year Published:
2024
Citation:
Chen, Rui, Clarissa Harris, Yin Sun, and Mengxue Li. "Strengthening HBCU Student Success through Data Science and Machine Learning Experimental Education in FANH." 1890 Center of Excellence for Student Success and Workforce Development (SSWD) Spring Symposium, May 2024. Tuskegee, AL.
- Type:
Conference Papers and Presentations
Status:
Accepted
Year Published:
2023
Citation:
Suchit Bapatla, Rui Chen, and Yin Sun. Image Classification of Fruits and Vegetables for Tracking Food Pantry Inventory. Artificial Intelligence for Future Agricultural Resilience, Management, and Sustainability (AIFARMS) Annual Conference, August 2023, UIUC, IL.
- Type:
Other
Status:
Published
Year Published:
2023
Citation:
ARD (Association of 1890 Research Directors) Updates. "Tuskegee increases AI use in land -grant programs," Dec 2023, Vol. 14, Iuse 12. www.umes.edu/ard
|