Validation of low-cost sensors for optimal insect protein production
According to FAO, the world population will reach 9.1 billion by 2050, and to feed that number of people, global food production will need to grow by 70%. For Africa, which is projected to be home to about 2 billion people by 2050, farm productivity must accelerate at a faster rate than the global average to avoid continued mass hunger. The current food challenges in Africa mainly stem from a rapidly growing population and low farm productivity which has been aggravated by climate change. In Kenya, the economic review of agriculture 2021 indicates that about 35% of the Kenyan population lack access to adequate food, with an estimated 2.1M people experiencing high levels of acute food insecurity.
But this is about to change. Adopting modern agricultural technologies, such as IoT sensors, have the potential to significantly boost African agriculture and make it more productive, profitable, sustainable, efficient and inclusive. In their 2018 report, Precision Agriculture and Food Security in Africa, authors Ncube, Mupangwa and French write that : “In Africa, significant benefits can be accrued by commercial and small-scale farmers through the implementation of precision agriculture practices. The benefits on a farm can be crop, economic and environment related.”
Building on nature-based frugal innovation
Although research has been conducted on using precision agriculture in Africa to increase crop yields and improve water use efficiency, nothing similar has been done in the rapidly growing sector of insect-farming, which is becoming a sustainable source of income and growth for small and medium scale farmers in Africa. Sanergy, a social enterprise based in Nairobi that manufactures high-quality insect-based animal protein (Black Soldier Fly larvae) from organic waste, has teamed up with Icipe (International Centre of Insect Physiology and Ecology) to provide a comprehensive proof of concept on the use of IoT sensor technology for optimal insect-based animal protein production.
Insect protein is a growth industry in Africa especially Kenya, where 4% of animal feed protein comes from insects. Two of Kenya’s rapidly growing sectors are livestock and aquaculture, which contribute to 12% of Kenya’s GDP. The size of the industry underpins an increasing demand in animal feed (~65,000mT/year!).
Beyond proof of concept, the IoT sensor technology will speed this growth, create new jobs and improve livelihoods across Africa. Additionally, this validation will also be relevant to the sanitation sector as Black Soldier Fly Larvae (BSFL) offer an eco-friendly way of managing human waste where sewerage systems aren’t available, especially in the crowded informal settlements of big cities in Africa. Any improvements in the BSFL production processes translate to a more effective and environmentally sustainable way of managing organic waste in Africa. If proven, the IoT sensor technology will therefore also allow Sanergy to drive sustainability in growing African cities through efficient management of organic and sanitation waste that would otherwise be left to decompose in landfills and dumpsites leading to generation of GHGs.
The set-up
The project foresees 4 sprints across 12 months, but this is subject to change contingent on the results of each stage. In the build-up to Sprint 1, we identified and triaged all stories and assumptions that contributed both directly and indirectly to the pilot. These were thoroughly reviewed and groomed, making the pilot roadmap clear and visible.
In Sprint 1, one lot of 11 experiments was performed. The experiments employed 5 types of sensors that were designed and developed by Wisense, a technology company that has been offering wireless/monitoring solutions in the agricultural industrial space since 2013. It is important to note that although research has been conducted and proven on using precision agriculture to increase crop yields and improve water use efficiency in Africa, nothing similar has been done in the rapidly growing sector of insect-farming, which is becoming a sustainable source of income and growth for small and medium scale farmers in Africa. The sensors tested in sprint 1 include:
- substrate moisture content sensors,
- substrate temperature sensors,
- ambient temperature sensors,
- ambient relative humidity sensors,
- ambient CO2 and air-flow velocity sensors.
These sensors were deployed both in large and small-scale set-ups and monitored over a period of one and half months. During this time, the sensors reliably relayed real-time data for the remote monitoring of Back Soldier Fly Larvae at various stages of their development in the substrate. This was an indication that the sensors can be reliably adapted, both at large and small-scale levels, to optimize production as well as intervene with precision to changing environmental factors.
With these initial results, the stars are definitely aligning for over 11,000 small-scale, 200 medium-scale and the few large-scale insect producers in Kenya who have the potential to serve the growing demand for insect-based animal feed in Africa. All these farmers will benefit from the learning of this pilot to advance their processes and optimize their production regardless of the size of their farms.
What we are taking away from Sprint 1
The lessons shared below indicate that the IoT sensor technology will be practical for the small-scale insect-based animal protein producers to replicate for three reasons:
(i) Acquisition of data through IoT sensor technology eliminates the need for manual data collection and analysis. The current industry practice involves time consuming manual paper-based data collection and difficult data analysis which often result in farmers completely abandoning any controls yet hoping for the best.
(ii) IoT sensor technology can widen the spectrum of data collection as it is possible to have multiple sensors plugged at the same time therefore providing extensive information about the processes.
(iii) The sensors are easy to deploy due to their modular design and their adoption should be easy as they require minimum interaction.
‘Herding cats’ — the challenges
While Sprint 1 results will definitely spark the attention of BSFL producers, challenges also existed:
(a) Internet reliability presented a key bottleneck. Intermittent loss of internet connection resulted in the loss of data as the sensor design under experiment did not incorporate internal data buffering for export when internet connection is re-established
(b) The distance of the sensors from the gateway (which communicates with the sensors over varying protocols and then translates that data into a standard protocol to be sent to the cloud) was also noted to be an issue, with a recommendation being made to have sensor nodes closer to the gateway for reliable connection. Therefore, additional gateways are recommended where sensors are greatly distributed.
What next?
Design improvements will be implemented, based on Sprint 1 findings and challenges, and additional key insect growth parameters will be monitored in Sprint 2. The IoT sensors will be integrated with an Enterprise Resource Planning system for better operational efficiency, communication and accurate forecasting. Pushing of sensor alarms to emails/SMS will also be tested in this next sprint.
You must have heard of technologies that were promoted as going to change the world forever. However, some of these technologies ended up being abandoned. From Sprint 1 results, we think IoT sensor technology will transform the world of insect farming and this could happen in a simple yet effective way. Do you want to imagine how this will happen or do you want to wait till IoT sensors in insect farming get into places you least expect? Look out for our next blogpost: The future of insect farming is happening now.
