2018 & 2019 WinnerRevealing informal food flows through free WiFi
The Inspire Challenge is an initiative to challenge partners, universities, and others to use CGIAR data to create innovative pilot projects that will scale. We look for novel approaches that democratize data-driven insights to inform local, national, regional, and global policies and applications in agriculture and food security in real time; helping people–especially smallholder farmers and producers–to lead happier and healthier lives.
This proposal was selected as a 2018 pilot project and 2019 scale-up runner up project, with the team receiving a total of 225,000 USD to put their ideas into practice.
Revealing informal food flows through free WiFi
Despite being the main source of fresh, convenient, and affordable food for 80% of Hanoi’s population, food flows within traditional markets remain largely invisible due to a lack of tracing systems and environmental conditions which make traditional tracking approaches challenging.
By providing free internet to a series of wholesalers and markets in the Cau Giay and Dong Anh districts of Hanoi, Vietnam, this project will put in place the first pieces of tracking system that will characterize and monitor food flows between traders, retailers, and consumers.
The collected food flow data will allow for improved linkages among key traditional market actors and help identify better policy and planning options for improving distribution channels in ways that benefits under-resourced communities.
Additionally, research has found that 10-40% of traditional market food is contaminated with microbes or parasites which cause foodborne illnesses. As shoppers become increasingly concerned about food safety and large-scale retailers that can offer food safety certification expand rapidly, this project aims to equip traditional market actors with data that could prevent their marginalization through urban policy decisions that may favor organized retailers, as well as improve the safety of traditional market goods.
To implement the project, partners CIAT and the General Statistics Office (GSO) of Vietnam will survey actors and track space and time data points on all devices within the range of the WiFi routers and signal amplifiers, whether connected to the internet or not.
The pilot system will run on three layers of data:
Nguyen Cong Thang | Email
Agriculture Department Vice-Director, General Statistics Office of Vietnam
Dr. Le Trung Hieu
Director of Agriculture, Forestry and Fishery Statistics Department
General Statistics Office of Vietnam
Louis Reymondin | Email
Scientist at CIAT
Huong Pham Thi | Email
Research Associate at CIAT
Dr. Thibaud Vantalon | Email
Deep-learning Expert at CIAT
Trong Van Phan | Email
Research Assistant at CIAT
Kien Tri Nguyen | Email
Data Manager at CIAT
Ysabel Anne Lee | Email
Communications Officer at CIAT
Stef De Haan
Layer One: Smartphone
Every smartphone has a unique media access control (MAC) address that the WiFi routers installed in the markers will use to identify how many MAC addresses visit the markets over time, how many return to the market and how often, and how the two markets differ on these metrics. This data can be collected even if the smartphone is not connected to the WiFi network.
Layer Two: WiFi
When a smartphone user connects to the free WiFi, they will be prompted to answer a series of questions depending on their user type (vendor, customer, etc.). For example, a user that identifies as a vendor will be asked questions regarding sales of specific commodities which will allow for sales to be characterized across time and space.
Layer Three: Surveys
To validate findings in Layer One and Two, “flash” surveys will be conducted using standard survey instruments.
Step by step
Scoping study for markets
The traditional market system in Vietnam is quite diverse, divided in three types (type one: more than 400 stores, type two: more than 200 stores, and type three: less than 200 stores).
A scoping study on markets was conducted in two districts in order to better understand market types and their management structures.
Local partners, local authority, and market management boards gathered for a kick-off meeting and get feedback on research using WiFi, which is considered a new approach in the area.
Five specific markets where the project will provide free WiFi and collect data were selected. Potential implementation constraints and solutions were discussed.
Router setup and WiFi installation
Data collection is carried out during this time.
The analysis expects to provide detailed network analysis of food flows at different scales for different markets as well as recommendations to policy makers related to market development.
US$125K scale-up grant
The project was awarded a 2019 Inspire Challenge Scale-up grant of US$125K at the third annual convention of the CGIAR Platform for Big Data in Agriculture, 16-18 October 2019.
The first project of its kind in Vietnam
“The research project on food flows in traditional markets using the WIFI system in order to trace food origins is the first project to be implemented in Vietnam,” says Dr. Le Trung Hieu, Director of Agriculture, Forestry and Fishery Statistics Department at the General Statistics Office of Vietnam.
“The results of the project are a very important source of information to help the management authorities better understand shopping habits of the local people, provide insights on main food supply channels in traditional markets (wholesale and retail markets), help policy makers have solutions to strengthen food safety control and plan safe agricultural production areas to meet the needs of the people, and control transporting means to ensure food and traffic safety and traffic safety.”
Hear more from Dr. Le Trung Hieu in this video interview:
The team is conducting a review of project progress to understand strengths and weaknesses of the design and adjust accordingly.
Expanding the team
Two graduate students will join the team: one for data analysis (from the University of Wageningen) , one for machine learning analysis of Layer One data (from the University of Applied Science of Western Switzerland).
A workshop will be held with the GSO, the Ministry of Agriculture of Vietnam, and a national university to discuss the survey aim.
The GSO and The Ministry of Agriculture are interested in market price data and how prices fluctuate. However, reliable price data has proved difficult to obtain through the current surveys, and the team is evaluating options to control for price data quality within survey responses.
Continued data collection
Based on the outcomes of the February workshop, the team will revise the survey, and continue data collection for eight months.
Informing indicators and policy
Based on the data collected, CIAT and GSO will discuss how the system can inform existing official indicators and a policy brief.