The HVT Research Knowledge Exchange (RKE) on Thursday 20th October explored how long-term and hugely expensive transport developments can be supported by robust research and data, and highlighted innovative approaches to data collection that make relying on data an affordable possibility for low-income countries.
Held via webinar, the RKE featured presentations by Tom Russell of Oxford University’s Programme for Sustainable Infrastructure and Philip Krause, transport engineer and Project Manager at GoAscendal.
The session was opened by moderator Henry Kerali, former World Bank Country Director, who outlined the huge investments made in infrastructure, highlighting that in 2021, the World Bank had 162 active transport projects in its portfolio amounting to US$31 billion, of which approximately US $20 billion are for connectivity (inter-urban and rural.) And in the pipeline is US$18 billion over the next three years. Against this backdrop, policy makers and transport practitioners need to be informed to make smart decisions.
The first speaker, Tom Russell, is working as part of a team from Oxford and Southampton universities to use open data to map and model the impact of climate change on transport systems. The project, Decision Support Systems for Resilient Strategic Transport Networks in Low Income Countries aims to provide support tools to transport decision-makers in LMICs to enable them to prioritise interventions that will deliver sustainable and long-distance transport networks.
Focusing on the issue of resilience, Tom outlined the problem of flooding and other climate hazards causing direct damage to road and rail networks as well as wider disruption to economies and societies. His team carried out analysis in four countries in East Africa, predicting an increase in expected damages in a given year by using future scenarios. In the current climate with current conditions there are considerable damages, he said. In the future in a medium emissions scenario that will increase, while in a higher emissions scenario, as floods become even more intense and frequent, damages will be even greater.
“Roads and rail are crucial to economic development as these countries urbanise and develop their economies, but we’ve also got to contend with changing climate which is imposing increasing risks to the infrastructure networks,” he said.
Tom’s team assess the risks posed by flooding, consider risk reduction and the cost of various interventions and how they could provide benefits. To do this they use global open data sources, in particular the World Resources Institute Aquaduct product, which features global flood hazard maps, so shows the likelihood of floods now and in the future with different climate change scenarios. Data sets on the networks give good coverage of the road and rail networks within the study countries so the team at Oxford and Southampton can see open street map roads in great detail.
“Plugging that together into the models and tools that we’ve developed has allowed us to produce some open source analysis and open data results”, he said.
Tom shared a web based tool of their analysis in the form of an interactive map to view the data sets they’d brought together and the risk, resilience and adaptation results that they’d calculated. In a live demo, he showed both open street map data and flood hazard data, illustrating how the tool can be used as an indicator to focus efforts and to offer different options according to various scenarios and cost benefit ratios.
Questions following Tom’s presentation included the reliability of open source data in low income countries, which Tom said was fairly good, adding that continual campaigns by humanitarian organisations to improve street coverage drives data collection and improvement all around the world.
Presenting the HVT-funded Africa Urban Mobility Observatory (AUMO), Philip Krause then outlined the work done by the Go Ascendal/Go Metro team.
“ Our focus was on developing an observatory ecosystem in which a range of data collection solutions was developed based on telecommunications technologies.”
The aim was to try and reduce the – usually prohibitive – cost associated with mobility data collection, one of the reasons why sources of data in Sub-Saharan Africa are often lacking.
Under the HVT project six research cities were identified: Blantyre, Gaborone, Kigali, Kinshasa, Lagos and Maseru and a list drawn up of suitable indicators on which to collect data. A range of data collection instruments was then developed using existing technologies as far as possible. In parallel the team developed a web data platform.
Data collected by these different methods was then pooled into data sets which could be expanded by adding additional cities, so providing support to transport decision-makers with limited budgets to address the mobility needs of people in their respective cites.
Philip described four different collection methods:
- Conventional field surveys, where enumerators would go to busy intersections with a tablet or smart phone and ask multiple choice questions;
- USSD surveys, a communication protocol compatible with all GSM devices, which made use of marketing SMSs and small monetary incentives;
- WhatsApp surveys, and
- User Movement Analytics (UMA) surveys, which involved tracking on user’s devices.
The number of surveys completed on the latter two methods was very low which contributed to poor final results. In a final cost comparison, the field surveys proved to be the least expensive across the six cities, although Philip noted that WhatsApp and USSD surveys could be cheaper had people completed them right through to the end so could provide a potential source of data going forward, especially given mobile network operators’ reluctance to share their big data.
Questioned about the challenges in ensuring data is being used effectively for transport development in LICs, Philip cited political will and capacity and described how Kigali, Rwanda, is making use of data provided through a parallel project run by UN Habitat and ensuring that their academic institutions can support similar types of data collection and analysis.
Tom highlighted engagement, time, resource and capacity within the network of infrastructure operators, different sections of government, funders and donor organisations, saying it needs to be well supported and stitched together. He said:
“It’s about getting ourselves further out of the picture, partly by making our tool easier to use so easier to transfer, and partly by putting the resource in the hands of the people who are closer to the action.”