The Institute for Fiscal Studies (IFS) was launched with the principal aim of better informing public debate on economics in order to promote the development of effective fiscal policy. It is Britain’s leading independent microeconomic research institute and covers subjects from tax and benefits to education policy, from labour supply to corporate taxation.
About the project
The IFS wanted to understand the relationship between commuting and the gender pay gap. This study was created after IFS figures showed the existence of a 'gender commuting gap'. The data showed that men spend longer commuting to work than women. The report shows the ‘gender commuting gap’ starts to widen after the birth of the first child in the family and continues to grow for around a decade after that. This bears a striking resemblance to the evolution of the gender wage gap.
Calculating travel times between thousands of locations
As part of the project, the IFS needed to measure the travel times between employment opportunities and different locations. This required TravelTime to calculate a huge volume of requests as they needed to do 40,000 origins to 40,000 destinations. This is a total of 1.6 billion routes needed. They also wanted to calculate the route using driving and public transport. This means that TravelTime Analytics calculated a total of 3.2 billion journey times.
The end result was a large volume of CSV files calculating the travel times from a single origin to every destination, using both driving and public transport. The total volume of data required was 120GB. The data was processed and returned to the client in under 1 working week.
Journey times were calculated assuming the user arrived by 8.45am. When using public transport it understands that the user must walk to the first stop, wait for the next timetabled service, take the journey and exit the relevant station. This is repeated if the user must make transport connections.
Why use TravelTime?
- No development skills needed - we handle all the code and output easy to understand CSV files.
- We can process large volumes of origin-destination pairs.
- Price - Google's price per element is $0.004 USD when buying 100,001–500,000. (Pricing for requests over 500k not available on the website). If using this pricing model, this job would've cost $12.8 million, making this type of analysis on Google not financially possible.