Mobile phones are used for water quality testing and for finding the nearest safe water sources in Tanzania. Photo by: mWater / DivatUSAID
/ CC BY-NC-ND
The “internet of things” is an emerging class of technologies which use machine-produced data, most commonly from sensors, communicated via connectivity technologies such as mobile phones, to gather real-time information, which can then be analyzed and used to improve processes and decision-making.
Internet of things has taken off in the last few years in developed countries, for example wearable sensors in watches and bracelets to measure vital signs and count steps and progress towards products such as self-driving cars.
While there are examples of internet of things applications in developing country settings, they are far less common and the majority are focused around the so-called smart city movement; a new model of urban design which integrates information and technology to build and run a city more sustainably and efficiently.
But not always. For example, installing low-cost sensors in water pumps in rural villages across sub-Saharan Africa to monitor the pump’s functionality so that repairs can be carried out more quickly, is one example of how internet of things technology can improve service delivery in developing countries.
The internet of things concept is surrounded by hype with experts predicting economic impacts of between $3.9 trillion and $11.1 trillion per year by 2025 according to consultants at McKinsey. Furthermore, the number of devices is predicted to grow enormously over the next few years, with 10 billion more internet connected devices coming online by 2019, and out of these 70 percent will be internet of things connections.
Devex spoke to John Garrity, USAID’s senior connectivity adviser, who sees huge potential for internet of things to improve development beyond “smart cities”.
Here are the highlights of the conversation.
How do you define the internet of things?
There are a lot of different definitions of internet of things out there. The one I use defines the internet of things as being present when you have digital communication between two entities where either machine-generated information (usually from sensors) is transmitted to another machine or person, or a person is sending information to a machine to conduct an action, such as closing the garage door.
What are the main opportunities for internet of things to enhance development?
I see four main channels by which internet of things can further development — through improving service delivery, monitoring and evaluation, and research, and by enabling communities to collect localized data for their own contexts.
1. Service delivery.
There is a lot going on in the field but to pick a few examples; how internet of things can be used in water, sanitation and hygiene interventions is a really interesting space. USAID is supporting a project which uses internet of things technology to improve water service delivery through the implementation of water flow sensors in East Africa, and we are also part of a project which uses water level sensors to monitor boreholes.
There is also a dimension around using sensors for behavior change around sanitation. For example, sensors can be used to measure whether behavior change has actually led people to use latrines (sensors on latrine doors) and wash their hands (water flow sensors). Mercy Corps Indonesia recently partnered with Portland State University’s SWEETLab to pilot this.
Internet of things also has strong applications in the resiliency and humanitarian space. For example, the Red Cross has been installing low-cost networks of fire sensors in informal settlements where fast moving fires have been a problem in Nairobi and Cape Town. The sensors can tell when a fire breaks out and inform close by households and the municipal authorities.
Natural resource management and specifically the protection of rainforested areas and wildlife can also be enhanced by internet of things technology. Conservationists use acoustic sensors to listen for unnatural sounds which could be associated with poaching or logging.
More examples can be found in the joint U.N. International Telecommunication Union and Cisco report, Harnessing the Internet of Things for Global Development, published in January of this year.
2. Internet of things and M&E.
A lesser discussed application of internet of things technology is its potential to improve monitoring and evaluation of development projects.
For example, the Global Alliance for Clean Cookstoves, supported by the U.S. government, has been using sensors to monitor in real time whether clean cookstoves, designed to reduce household air pollution, were actually being used. Without the sensors, M&E staff would need to rely solely on survey data, which is usually collected one year after the project starts, in order to build a picture of how the stoves were being used. This data can be unreliable and subject to recall biases from the respondents. However, with data from the sensors, USAID staff were able to draw a more complete picture of how, when, and for how long the stoves were used.
However, it’s important not to see sensor data as an alternative mode of data collection, it needs to be used alongside in-person survey data, since sensor data has its own biases, for example, are people prompted to use the stove more frequently than they would have normally because they know the sensor is there?
3. Research and policy formulation.
There is huge potential for the internet of things to improve research and currently there is a lot of activity from academic institutions around training researchers from developing countries to use sensors to improve their own research activities.
On the policy side, we see examples of where data being collected from the internet of thing is being used to help policymakers make decisions about allocation. For example, data on village hand pumps can help inform municipalities about where to install future pumps based on which villages are seeing the most use and demand.
4. Local data for local contexts.
Normally when large amounts of data are collected this is done by major public entities at the national level and then fed down to communities, whether it’s information about health, education or the environment, for example.
However, sensors and the internet of things allow communities to collect data for their local contexts. For example, information about local weather or flood monitoring. The DAI Hidrosónico project is working with flood-prone villages in Honduras to use sonar sensors, which are mounted on bridges, to monitor river water levels. When levels get dangerously high, villages are alerted so they can evacuate in time.
Through this we can see a role for the internet of things in rural and per-urban communities, so it’s not just about “smart cities” but also “smart towns” or “smart villages.”
What are the major challenges for internet of things and development?
There are technical challenges around bandwidth, connectivity, power supply, and scaling sensors. Technical capacity is also a major issue, whether internal to the organization trying to deploy sensors or challenges on the ground about how communities maintain them. These technical challenges will only be solved by more players using sensors and appreciating how real time data can improve development operations.
There are also challenges on the policy side related to cross-border data flows which is becoming difficult as more and more countries are mandating that data collected has to stay within the country.
Issues around open data are also coming up — we have seen instances where countries have tried to close down projects because the data being gathered was shining a negative light on certain things.
Privacy and security are other issues which will need to be tackled — how do we make sure people have the right agency and understand what information is being collected, and how do we do this across different countries which have different policies?
On security, the issue is about one way communication — devices send information but there is no way to communicate with the device itself. This was seen in the recent distributed denial-of-service attack against DNS service provider Dyn, which saw hackers hijack internet of things devices including video cameras which had lax security. Authorities were unable to communicate with the cameras to shut them down.
Every time you connect a new device to the internet, you’re adding another door through which to conduct malicious activity. To tackle this, we need to make sure the network is secure and that it is authenticating and also de-authenticating devices where appropriate.
What do you see as the future of internet of things and development?
It’s still very early days in terms of the internet of things ecosystem and we are seeing that even in advanced countries. This immature ecosystem means there just aren’t many solutions you can buy off the shelf and immediately deploy. However, more and more companies and social enterprises are filling this space and it’s building momentum and the ecosystem.
Interestingly, I’ve noticed that a lot of the innovative internet of things and development work is coming out of international NGOs, academia, and social enterprises. It’s not the large multilaterals. Going forward, I hope to see more mainstreaming of these solutions from people working in large multilaterals.
Read more international development news online, and subscribe to The Development Newswire to receive the latest from the world’s leading donors and decision-makers — emailed to you FREE every business day.