Search for: All records

Fleet maintenance is the process fleet manager utilizes to manage fleet and asset information from acquisition to disposal. It helps the companies reduce costs, improve efficiency and safety. Second Harvest of Metrolina Food Bank (SHMETROLINA) distributed over 70 million pounds of food and household items to approximately 800 partner agencies in 2019. With the critical need for transportation for distribution, a vehicle experiencing downtime will disrupt scheduled routes to partner agencies and increase repair costs for SHMETROLINA. This research developed an interactive dashboard using R shiny to help non-profit food bank fleet managers make informed decisions for effective fleet maintenance operations and support the food bank operations to meet hunger needs. The dashboard consists of the visualizations of the maintenance cost, mileage, and operational cost for individual and fleet vehicles from the data collected by the food bank. 

North Carolina is labeled as the 10th hungriest state in America, with almost 1 in 5 children in North Carolina facing hunger regularly. Based on these numbers alone, childhood hunger is an important issue that needs to be addressed. This paper focuses on the thirty-four counties serviced by the Food Bank of Central and Eastern North Carolina. Each county is mapped out, showing the percentage of children receiving free or reduced meals. These numbers are then compared to the number of Weekend Power Packs, Kid Cafes, and School Pantry locations per county. This project illustrates the strength of visual analytics for decision making 

University food pantries have been opened on-campus to reduce food insecurity among students. However, they are relatively self-governed and receive a limited amount of support. This study aims to resolve the inventory management issue at university food pantries using the combination of the Model-View-Controller (MVC) software pattern and data visualization. As a result, a foundation is established to predict the food choices of clients and to manage food waste effectively. The benefits of visualization on decision making through the use of a resourceful inventory system are outlined. The focus of this research is on university pantries, in particular, the Aggie Source Food Pantry at North Carolina Agricultural & Technical State University. A sample of 50 clients' food choices was acquired from picklists for June and July of 2019. The inventory tracking system implemented is a client-server mobile application used for data collection. Data visualization was applied to evaluate the food donations and distributions. Students preferred essential foods (e.g., pasta, canned vegetables), over unhealthy foods (e.g., Pop-Tarts, cookies). The data consisted of 338 pounds of distributed food and almost 2,473 pounds of donations. Data was simplified into comprehensive visual diagrams. The MVC structure established a program interface that grouped application functions and managed data objects. The solution allowed the staff to understand pantry status and the trends in production. It is anticipated that the prototype will be implemented in the daily activities of the pantry. 

North Carolina is the third most hurricane-prone states in the US. In 2018, Hurricane Florence caused a lot of damages to households in North Carolina. The Food Bank of Central and Eastern North Carolina (FBCENC) serves 34 counties in North Carolina, and 22 of them were affected by Hurricane Florence. This research aims to investigate the impact of Hurricane Florence on the operations of FBCENC. We developed interactive dashboards to visualize food bank operational data and other relevant data and studied the trends and patterns of food distribution in three key stages: preparedness, response, and recovery. These dashboards enable food bank operations managers to explore and interact with the data with ease to explore the operational data at different stages, at different branch level, and on a different time scale (monthly, weekly, or daily). The impact on the operations of affected service areas vs. not affected areas could be investigated as well. The findings of this research will provide insight into how humanitarian relief agencies can better prepare for, respond to, and recover from the disruptions caused by hurricanes. 

Community Food Strategies provides guidance and tools to local food networks in North Carolina to empower the creation of equitable policy change at all levels. Their vision is the realization of an equitable food system that improves the quality of life for all. In Davidson County, the recently formed Local Food Network, created under the guidance of Community Food Strategies, is working to create a more equitable and sustainable food system and is evaluating how best to apply their resources toward this end. The county is largely rural, and rural areas tend to have lower food access. Two cities, Lexington and Thomasville, have experienced economic depression over the past three decades with the exodus of furniture manufacturing overseas and as such, both cities have communities that suffer from poverty and food insecurity issues. In this research, we use visual analytics and spatial analysis to build a survey and analysis of local food availability for the county. Visual analytics is increasingly being applied to applications like this to increase understanding of large datasets and improve decision making. This research provides a base analysis for the Local Food Network to present the state of the local food system to potential partners who can help create policies that will increase equity and availability of food to everyone in the county. 

We present an analysis of factors contributing to the annual level of donation of sweet potatoes in 2010-2016 to a North Carolina food bank. Our approach follows that of Su et al., who used the Analytic Hierarchy Process (AHP) and Dempster-Shafer theory (DST) to assess annual grain security in China for 1997-2007. We first identified the indices (or factors or criteria) that influence the level of donation and their “directions:” positive (the more the better), negative, or non-directional (average is best). We divided the range of each index into degrees (intervals) then applied AHP to get weights for the indices. To apply DST, we defined a frame of discernment that would generate focal elements that could be assigned to degrees of the indices. Then, using the index weights, we defined a BPA (basic probability function) for each year. Since for each year we had multiple pieces of possibly conflicting evidence, we used Dempster’s rule to combine each BPA with itself several times. In the resulting BPA, the focal element with greatest mass was taken as the prediction for the donation level for that year. We partitioned the range of the donation data into degrees to compare observations with the focal elements in the BPA. Predicted donation degrees matched observed degrees reasonably well if degree boundaries are well chosen. Analysis of apparent anomalies suggested a more sophisticated approach and the need to involve other information sources. This approach allows one to experiment in a principled way (and without assumptions about probability distributions) with the relative importance of the multiple factors that affect the predicted quantity and so to understand how these factors together contribute to that quantity. It is suggested for gaining preliminary insight, which may be exploited in the application of more rigid analytic techniques. 

Predictive modeling of a rare event using an unbalanced data set leads to poor prediction sensitivity. Although this obstacle is often accompanied by other analytical issues such as a large number of predictors and multicollinearity, little has been done to address these issues simultaneously. The objective of this study is to compare several predictive modeling techniques in this setting. The unbalanced data set is addressed using four resampling methods: undersampling, oversampling, hybrid sampling, and ROSE synthetic data generation. The large number of predictors is addressed using penalized regression methods and ensemble methods. The predictive models are evaluated in terms of sensitivity and F1 score via simulation studies and applied to the prediction of food deserts in North Carolina. Our results show that balancing the data via resampling methods leads to an improved prediction sensitivity for every classifier. The application analysis shows that resampling also leads to an increase in F1 score for every classifier while the simulated data showed that the F1 score tended to decrease slightly in most cases. Our findings may help improve classification performance for unbalanced rare event data in many other applications.