A senior data analyst at Lockheed Martin, Kremer creates systems to efficiently gather, present, and automate data, enabling clear insights into performance and streamlining processes. But after hours and on weekends, he is all about tending to a 10’ by 20’ backyard garden that, this year, produced not only abundant bell peppers, asparagus, Brussels sprouts, and shallots but also close to 100 pounds of tomatoes.
A fresh food enthusiast, Kremer became aware of and concerned about low-food access areas in Dayton while studying industrial engineering at the University of Dayton. For his Systems Engineering capstone project, he designed a “Garden Management System” that promises to increase harvest yields for urban community gardens and farmers’ markets.
Tell us more about what inspired this project.
As a student at the University of Dayton, I noticed that the city has numerous areas where access to fresh produce is not available via local grocery stores. After speaking with urban garden organizers there, it was evident that technological improvements could alleviate some of the burdens they face when managing a garden and could improve produce yields for local residents who rely on gardens for their families.
What was your solution? How does it do the job better than systems already in use?
The solution I developed is called “The Garden Management System.” It offers integrated capabilities for providing irrigation to a garden, managing the distribution of fertilizer to plants, controlling weeds, deterring animal pests from entering the garden, and providing real-time data insights into the health and status of the garden.
The Garden Management System offers features and capabilities that go beyond the current smart garden solutions on the market. For example, pest deterrence would be enhanced using embedded classification models to determine the types of animals entering the garden and sending alerts to the owner. Irrigation and fertilizer deployment would be based on the optimal soil moisture and nutrient levels for specific plant types throughout the garden. Data insights would also provide the owner with suggested planting dates, keep track of germination schedules, and estimate harvest periods by plant.
How did you use systems engineering to solve this problem?
I followed the typical iterative systems engineering process. Before identifying any solution, I researched extensively and interviewed numerous stakeholders to accurately define the problem and understand the system’s needs. This included research into FDA food deserts, urban garden efficiency reports, current state reviews of local food bank gardens, manual garden task reviews, and interviews with local gardeners.
Once I identified what the project should accomplish, I then moved towards defining a set of requirements to meet the operational needs of managing a garden and displaying key data insights to the garden’s owner.
The system characteristics and operational capabilities of the Garden Management System were also outlined in a concept of operations (CONOPS) document, along with scenarios that included inputting plant layouts, scaring away pests, receiving data insights for delayed germinations, and deploying irrigation to areas of a garden. I then used these scenarios and concepts to define the activities and interactions of the garden system through a Functional Analysis. Examples of functions outlined include managing soil conditions, removing weeds, storing data, and controlling external garden threats. Outlining system functions led me to analyze the physical domain and the types of hardware/software that would be required to physically construct the system and perform those functions. For example, ultrasonic transmitters were identified as a way to scare away wildlife pests. Since there are quite a few different approaches and components that could be used as physical parts, I performed trade studies to compare alternatives and make selections. Additionally, the complex nature of the system, and the long list of outlined requirements, required me to create a test and evaluation plan to ensure all capabilities of the garden system were being validated and verified. All steps in the systems engineering process were iterative, allowing the Garden Management System to be further refined throughout its development.
Why is this work important? Do you see your system being implemented in real life?
This work is important since it offers a real solution for individuals who rely on community and local gardens for their fresh produce. A surprising percentage of the United States population lives in areas where gas stations and convenience stores are their only options for food. The Garden Management System promises to not only boost harvest yields but also to improve the chances of an urban garden’s survival due to the automation of manual work normally conducted by hard-to-find volunteers.
I believe this type of system can be implemented in real life and could greatly improve the quality of food individuals eat. It also improves the self-sufficiency of residents and reduces the carbon footprint involved with the supply chain of shipping non-local produce to big-chain stores.
Are you planning to continue to work on this?
As an avid gardener, I am continuing to refine this concept and manifest the idea into a working solution. My home garden now contains elements that I have built for the irrigation, fertilizer, and pest control components of the Garden Management System. As I continue to build and refine my system, I hope to partner with local food banks and community gardens to implement autonomous solutions into their manual gardening processes.