Development of internet of things framework of monitoring poultry housing in the poultry farming sector in Zimbabwe

Author

Chuma, Mufaro

Title

Development of internet of things framework of monitoring poultry housing in the poultry farming sector in Zimbabwe

Abstract

The use of latest technologies such as internet of things has enabled farmers to better care for their animals and produces more food and minimized cost of labor through the use of technology. Most farmers used traditional methods such as in person inspection to monitor bird welfare and its environmental conditions which was expensive. This research relates to the development of internet of things framework of monitoring poultry housing in the poultry farming sector in Zimbabwe. The study objectives were to design a model aimed at monitoring temperature, humidity and water levels in the poultry house, to develop an application or software which was linked on a mobile device to monitor temperature, humidity and water levels in the poultry house, to evaluate the performance of the developed model focusing on its accuracy, predictive power and its reliability. Its aim was to improve farm productivity, bird welfare and resource efficiency. The developed model comprised of three main components: the perception layer consists of various sensors such as the water level sensor, ultrasonic sensor and devices that measured parameters such as temperature, humidity and water levels. The network layer enabled communication between devices and transmitted information. The application layer processed and analyzed information and provided useful information to end users through a user friendly interface. This data was analysed to provide actionable insights and driven farmers to make decision. A system analysis was conducted to identify stakeholder needs, functional and nonfunctional requirements, and technical specifications for the IoT system. A pilot implementation was carried out to evaluate the systems performance, usability and impact on farm operations. The study explored the potential integration of machine learning algorithms and artificial intelligence to enhance data analysis and predictive modeling capabilities. After the model was developed, the model was tested for its performance focusing on its accuracy, predictive and reliability which was a success. The experiment that was carried out for five days and its success, failures, Mean Time between Failures, Mean Time to Repair, and Uptime and downtime percentages was calculated and results showed that:
1. Reliability
A system’s reliability is a measure of its ability to perform its intended function without failure.
With a mean time between failures of 60h and an uptime percentage of 98 % , the system appeared to be quite reliable. Because the observed and recorded temperature, humidity and water levels in the coop”s water system which was recorded was taken from both the dashboard and the blynk application showing real time data. For example on Tuesday between 12pm to 16:00 pm the temperatures ranged from 20 to 26 ºC and humidity ranged from 39 to 42 % and water levels in the coop”s water system ranged from 40 % decreasing to 10 %. This was because during the day temperature raised to 26 ºC and humidity to 42 % (optimum conditions), the chickens will be active as well and the food consumption will be high so they will be drinking water that’s why it dropped from 40 to 10 %. When the water level dropped below 15 % the model did sent a notification to the application flashing the blue light meaning that the water level has dropped and a refill is needed also a buzzer started producing a sound, notifying people who were around that water levels had dropped. Also on Wednesday the temperaturewas high at 30 degrees around 3pm the system responded very well to the condition which was considered to be high end of the comfortable temperature range for chickens and the system gave a notification on the application that ventilation was needed in the coop.
2. Maintainability
This refers to how a system can be easily repaired and maintained. With a Mean Time to Repair
of 1 hour, the system appears to be fairly maintainable;
3. Availability
This refers to the proportional of time a system is operational and available for use. With this system, a downtime percentage of 3.3 %, the system is highly available, with limited downtime.
The findings in this study contributed to the advancement of precision agriculture and informed
the development of tailored IoT solutions for poultry farming industry. The proposed system
framework was expected to help farmers to optimize resource usage, detect and mitigate potential risks and also to enhance farm productivity. However addition of other parameters on the model is recommended such as measure of illumination and number of chickens in the coop to be shown on the dashboard.

Date

June 2026

Publisher

BUSE

Keywords

Internet of Things (IoT)

Poultry farming

Precision agriculture

Smart farming

Supervisor

Professor Gongora, Ever