In farming operation, machinery represents a major cost; therefore, good fleet management can have a great impact on the producer’s profit, especially considering the increasing costs of fuel and production inputs in recent years. One of the tasks to be accomplished in order to improve fleet management is planning the path that the machine should take to cover the field while working. Information such as distance traveled, time and fuel consumption as well as agricultural inputs wasted due to off-target-application areas are crucial in the path planning process. Parameters such as field boundary size and geometry, machine total width as well as control width resolution present a great impact on the information necessary for path planning. Researchers around the world have proposed methods that approach specific aspects related to path planning, the majority addressing machine field efficiency per-se, which a function of total time spent in the field as well as effective working time. However, wasted inputs due to off-target-application areas in the maneuvering regions, especially in oddly shaped agricultural fields might be as important as field efficiency when it comes down to the total operation cost. Thus, the main purpose of this research was to develop a path planning method that accounts for not only machinery field efficiency, but also the supply inputs. This research was accomplished in a threefold approach where in the first step an algorithm for computing off-target application area was developed, implemented and validated resulting in a computational tool that can be used to evaluate potential savings when using automatic section control on agricultural fields of complex field boundary. This tool allowed accomplishment of the second step, which was an investigation and better understanding of field size and shape as well as machine width of the effects on off-target application areas resulting in an empirical method for such estimations based on object shape descriptors. Finally, a path planning algorithm was developed and evaluated taking into consideration the aspects of machine field efficiency as well as off-target application areas.

Biochars have been created and characterized from a variety livestock manure biomass including poultry, dairy, and swine. However, no research has been conducted on the physical and chemical makeup of biochar pyrolyzed from horse muck. Two horse muck derived biochars containing either straw (HS) or woodchip (HW) bedding were pyrolyzed at 700°C and characterized for their physical and chemical properties. Tests revealed both biochars had high alkalinity, moderate specific conductivity, and low surface area as compared to other biochars in the literature. HS contained more mineral structures than HW. Scanning electron microscopy presented differences in particle shape, size, and presence of xylemic structures. The chemical makeup of both biochars was similar, while HW contained about 23% more C than HS while HS contained more calcium and magnesium. The effect of biochar-amended soils on the transport of two Escherichia coli isolates was studied using saturated soil columns. The results show that HW significantly reduced the transport of isolate SP2B07 over the Soil control, and reduced the transport of isolate SP2B07 more than isolate SP1H01. Horse muck biochars may show promise in reducing bacterial transport though agricultural soils.

This project aimed investigated cellulase in-situ production for large-scale on-farm production of lignocellulosic biofuel. Cellulase activity and glucose released by T. reesei with corn stover and wheat bran as co-substrates for solid state cultivation (SSC) were examined. Co-cultivation has previously increased T. reesei cellulase, but corn stover and wheat bran have not been co-cultivated (Dhillon, Oberoi et al. 2011). This work compared cellulase activity and glucose concentration of corn stover co-cultivated with 0-40% wheat bran in high solids. Samples with at least 20% wheat bran exhibited increased cellulase activity. However, the average glucose concentration without wheat bran was 3.29 g/L compared to 16.7 g/L with wheat bran. Glucose released by T. reesei on pretreated corn stover with 0-40% wheat bran was compared at the optimal temperatures for fungal growth and for cellulase activity after SSC. Previous research has rarely used cellulase from SSC to hydrolyze lignocellulose. Following SSC of T. reesei at 30°C for seven days, samples were warmed to 50°C for five days. Glucose concentration increased to 12.1 and 32.7 g/L for samples with and without wheat bran. This strategy could reduce lignocellulosic fuel production costs by eliminating need for commercial cellulase and is promising for efficient cellulose hydrolysis.

Agricultural producers consider utilizing multiple machines to reduce field completion times for improving effective field capacity. Using a number of smaller machines rather than a single big machine also has benefits such as sustainability via less compaction risk, redundancy in the event of an equipment failure, and more flexibility in machinery management. However, machinery management is complicated due to logistics issues. In this work, the allocation and ordering of field paths among a number of available machines have been transformed into a solvable Vehicle Routing Problem (VRP). A basic heuristic algorithm (a modified form of the Clarke-Wright algorithm) and a meta-heuristic algorithm, Tabu Search, were employed to solve the VRP. The solution considered optimization of field completion time as well as improving the field efficiency. Both techniques were evaluated through computer simulations with 2, 3, 5, or 10 vehicles working simultaneously to complete the same operation. Furthermore, the parameters of the VRP were changed into a dynamic, multi-depot representation to enable the re-route of vehicles while the operation is ongoing. The results proved both the Clarke-Wright and Tabu Search algorithms always generated feasible solutions. The Tabu Search solutions outperformed the solutions provided by the Clarke-Wright algorithm. As the number of the vehicles increased, or the field shape became more complex, the Tabu Search generated better results in terms of reducing the field completion times. With 10 vehicles working together in a real-world field, the benefit provided by the Tabu Search over the Modified Clarke-Wright solution was 32% reduction in completion time. In addition, changes in the parameters of the VRP resulted in a Dynamic, Multi-Depot VRP (DMDVRP) to reset the routes allocated to each vehicle even as the operation was in progress. In all the scenarios tested, the DMDVRP was able to produce new optimized routes, but the impact of these routes varied for each scenario. The ability of this optimization procedure to reduce field work times were verified through real-world experiments using three tractors during a rotary mowing operation. The time to complete the field work was reduced by 17.3% and the total operating time for all tractors was reduced by 11.5%. The task of a single large machine was also simulated as a task for 2 or 3 smaller machines through computer simulations. Results revealed up to 11% reduction in completion time using three smaller machines. This time reduction improved the effective field capacity.

The ability to quantify soil moisture spatial variability and its temporal dynamics over entire fields through direct soil observations using remote sensing will improve early detection of water stress before crop physiological or economic damage has occurred, and it will contribute to the identification of zones within a field in which soil water is depleted faster than in other zones of a field. The overarching objective of this research is to develop tools and methods for remotely estimating soil moisture variability in agricultural crop production. Index-based and machine learning methods were deployed for processing hyperspectral data collected from moisture-controlled samples. In the first of five studies described in this dissertation, the feasibility of using “low-cost” index-based multispectral reflectance sensing for remotely delineating soil moisture content from direct soil and crop residue measurements using down-sampled spectral data were determined. The relative reflectance from soil and wheat stalk residue were measured using visible and near-infrared spectrometers. The optimal pair of wavelengths was chosen using a script to create an index for estimating soil and wheat stalk residue moisture levels. Wavelengths were selected to maximize the slope of the linear index function (i.e., sensitivity to moisture) and either maximize the coefficient of determination (R2) or minimize the root mean squared error (RMSE) of the index. Results showed that wavelengths centered near 1300 nm and 1500 nm, within the range of 400 to 1700 nm, produced the best index for individual samples; however, this index worked poorly on estimating stalk residue moisture. In the second of five studies, 20 machine learning algorithms were applied to full spectral datasets for moisture prediction and comparing them to the index-based method from the previous objective. Cubic support vector machine (SVM) and ensemble bagged trees methods produced the highest composite prediction accuracies of 96% and 93% for silt-loam soil samples, and 86% and 93% for wheat stalk residue samples, respectively. Prediction accuracy using the index-based method was 86% for silt-loam soil and 30% for wheat stalk residue. In the third study, a spectral measurement platform capable of being deployed on a UAS was developed for future use in quantifying and delineating moisture zones within agricultural landscapes. A series of portable spectrometers covering ultraviolet (UV), visible (VIS), and near-infrared (NIR) wavelengths were instrumented using a Raspberry Pi embedded computer that was programmed to interface with the UAS autopilot for autonomous reflectance data acquisition. A similar ground-based system was developed to keep track of ambient light during reflectance target measurement. The systems were tested under varying ambient light conditions during the 2017 Great American Eclipse. In the fourth study, the data acquisition system from the third study was deployed for recognizing different targets in the grayscale range using machine learning methods and under ambient light conditions. In this study, a dynamic method was applied to update integration time on spectrometers to optimize sensitivity of the instruments. It was found that by adjusting the integration time on each spectrometer such that a maximum intensity across all wavelengths was reached, the targets could be recognized simply based on the reflectance measurements with no need of a separate ambient light measurement. Finally, in the fifth study, the same data acquisition system and variable integration time method were used for estimating soil moisture under ambient light condition. Among 22 machine learning algorithms, linear and quadratic discriminant analysis achieved the maximum prediction accuracy. A UAS-deployable hyperspectral data acquisition system containing three portable spectrometers and an embedded computer was developed to classify moisture content from spectral data. Partial least squares regression and machine learning algorithms were shown to be effective to generate predictive models for classifying soil moisture.

This dissertation investigated issues surrounding grain harvest and transportation logistics. A discrete event simulation model of grain transportation from the field to an on-farm storage facility was developed to evaluate how truck and driver resource constraints impact material flow efficiency, resource utilization, and system throughput. Harvest rate and in-field transportation were represented as a stochastic entity generation process, and service times associated with various material handling steps were represented by a combination of deterministic times and statistical distributions. The model was applied to data collected for three distinct harvest scenarios (18 total days). The observed number of deliveries was within ± 2 standard deviations of the simulation mean for 15 of the 18 input conditions examined, and on a daily basis, the median error between the simulated and observed deliveries was -4.1%. The model was expanded to simulate the whole harvest season and include temporary wet storage capacity and grain drying. Moisture content changes due to field dry down was modeled using weather data and grain equilibrium moisture content relationships and resulted in an RMSE of 0.73 pts. Dryer capacity and performance were accounted for by adjusting the specified dryer performance to the observed level of moisture removal and drying temperature. Dryer capacity was generally underpredicted, and large variations were found in the observed data. The expanded model matched the observed cumulative mass of grain delivered well and estimated the harvest would take one partial day longer than was observed. Usefulness of the model to evaluate both costs and system performance was demonstrated by conducting a sensitivity analysis and examining system changes for a hypothetical operation. A dry year and a slow drying crop had the largest impact on the system’s operating and drying costs (12.7% decrease and 10.8% increase, respectively). The impact of reducing the drying temperature to maintain quality in drying white corn had no impact on the combined drying and operating cost, but harvest took six days longer. The reduced drying capacity at lower temperatures resulted in more field drying which counteracted the reduced drying efficiency and increased field time. The sensitivity analysis demonstrated varied benefits of increased drying and transportation capacity based on how often these systems created a bottleneck in the operation. For some combinations of longer transportation times and higher harvest rates, increasing hauling and drying capacity could shorten the harvest window by a week or more at an increase in costs of less than $12 ha-1. An additional field study was conducted to examine corn harvest losses in Kentucky. Total losses for cooperator combines were found to be between 0.8%-2.4% of total yield (86 to 222 kg ha-1). On average, the combine head accounted for 66% of the measured losses, and the total losses were highly variable, with coefficients of variation ranging from 21.7% to 77.2%. Yield and harvest losses were monitored in a single field as the grain dried from 33.9% to 14.6%. There was no significant difference in the potential yield at any moisture level, and the observed yield and losses displayed little variation for moisture levels from 33.9% to 19.8%, with total losses less than 1% (82 to 130 kg dry matter ha-1). Large amounts of lodging occurred while the grain dried from 19.8% to 14.6%, which resulted in an 18.9% reduction in yield, and harvest losses in excess of 9%. Allowing the grain to field dry generally improved test weight and reduced mechanical damage, however, there was a trend of increased mold and other damage in prolonged field drying.

Traditionally, burley tobacco has been harvested by hand because the green plant weight, volume, and leaf fragility make mechanical harvesting very challenging. This study examined possible ways to wilt a plant still standing in the field (termed ‘pre-wilting’) to reduce weight, volume, and leaf fragility. Several methods of pre-wilting burley tobacco plants in the field were explored including: root pruning, stalk girdling, freezing with liquid nitrogen, and burning. Experiments were conducted in three locations over three consecutive years during the tobacco harvest season. Leaf breaking angle, leaf moisture content and time-lapse photography were investigated as methods to quantify treatment effects on wilting. The time-lapse photography helped reveal that wilting was most prevalent during the late afternoon, and that wilted plants sometimes began to recover after more than five days, apparently due to root re-growth. Root pruning was the only mechanical means that caused witling reliably during the first two years of testing, and even then the results were somewhat inconsistent. During the third year, a high-clearance tobacco sprayer was modified with a hydraulically actuated coulter disc in order to root-prune a large number of subjects.

Fast-paced lifestyles result in consumers replacing traditional meals with on-the-go snack foods. In general, snacks are higher in saturated fats and simple sugars, and pose health concerns for consumers, which prompts the need for healthy nutritious alternatives to common snacks. Proso millet is a nutritious, and fast growing gluten free cereal. Spent grain (SG), the main by-product of brewing and distilling, contains high amount of protein and insoluble fibers. This study utilized proso millet and spent grain in the production of an extruded expanded snack, demonstrating their appropriateness as an ingredient in food production. The first objective of this study was to determine the effect of extrusion processing conditions (moisture, barrel temperature, screw speed) on the physical, chemical, and functional properties of proso millet based extruded snack. The second objective of this research was to determine the effect of distiller’s spent grain (DSG) addition level and particle size on physicochemical and functional properties of extruded snacks. The third objective of this research was to evaluate the effect of spent grain type and particle size on the physicochemical and functional properties of extruded snacks. Samples were produced using a co-rotating twin-screw 25:1 L/D laboratory scale extruder.

Ventilation and air movement are important aspects of animal agriculture and is frequently neglected in equine facilities. This paper discusses three different studies that examine different components of ventilation and air movement. One is a fan orientation study which examines how fans impact the stall environment, the second is a cooling study questioning whether forced air speed across a horse increases the rate of cool out after intense exercise, and the last is a survey examining ventilation, air quality, and health concerns in indoor arenas. The stall fan study took place over two summers with the goal of determining how the placement and orientation of different fans impacted the temperature within the stall, the air movement around the stall, and if the fans could provide fly control. Two barns with vastly different designs and natural ventilation properties were used. The barn in the first year had good natural ventilation, while the barn in the second year did not. Overall, the fans had little to no effect on reducing the temperature within the stalls, providing air movement throughout the stall, and did not produce sufficient air movement for fly control. After intense exercise such as a running a race, cross country, or participating in an endurance race, it is necessary to cool the horses and bring their vital signs back to resting ranges. The predominant method for this is hand walking or drenching and scraping the horses until the heart rate, respiration rate, and rectal temperature have returned to an acceptable level. The cooling study sought to examine whether providing forced air speed across the horses increased their rate of cool out through placing a Bannon Tilted Belt Drive 42 in drum fan around the horses during the drenching period of their cool out process. Heart rate, respiration rate, and rectal temperature were all monitored throughout the cool out process and the rate of return to resting values of the vital signs was used to determine the effectiveness of the cooling techniques. The presence of the fan and the air speed across the horses tended to increase their rate of cool out after exercise with the fan blowing from the hindquarters towards their head provided the greatest increase. Finally, the indoor arena study included an online survey and site visits with the purpose of gathering information regarding indoor arenas. As this is an under-researched topic, the goal of this study was to establish common characteristics, identify problems or issues within the facilities, and any health concerns for the horses and humans who use the facilities. The information gathered in this study covered a multitude of topics including arena construction and design, arena usage, footing type, maintenance practices, environmental concerns, and potential health issues within the facilities. This study will serve as the framework to build future research studies to examine and rectify issues within the facilities and, ultimately, provide design recommendations for building or retrofitting indoor arenas to mitigate or eliminate concerns.

Variable-rate technologies for sprayer applications stand to increase efficacy by ensuring the right amount of chemical is applied at the right location. However, external environmental factors such as droplet drift caused by variable ambient condition, are not yet integrated into modern sprayer systems. Real-time wind velocity measurements can be used to control droplet spectra for reducing spray drift by actuating a variable-orifice nozzle. This work aimed to develop data processing methods needed to filter noise and remove vehicle speed from wind velocity measurements collected with an ultrasonic anemometer aboard a moving platform. Using a global navigation satellite system (GNSS), vehicle speed was calculated in the field and subtracted from apparent wind velocity for comparison to static measurements. Experiments under stationary and dynamic sensor deployments were used to develop an algorithm to provide instantaneous local wind velocity and to better understand the local spatiotemporal variability of wind under field conditions.