The UPS Environmental Sustainability Award: CitySprint

Award Type: 

Finalist for Business in the Community's UPS Environmental Sustainability Award 2019

CitySprint: Machine-learning algorithms at heart of low-carbon delivery strategy 

CitySprint has a fleet of more than 5,000 couriers out on the roads. Reducing its carbon footprint is critical for the same-day delivery company, especially at a time when the pressure is on to tackle air pollution. 

While its team of couriers delivers the parcels, it is the cloud-based technology that drives the efficiency behind the scenes. With more than 15 years of historical delivery data, the business is using machine-learning algorithms to consolidate its work. This reduces delivery mileage, times and empty return journeys, helping to cut emissions and get traffic off the roads. The technology automatically allocates jobs to the best placed courier, making use of the right vehicle, for the right parcel, at the right time. 

CitySprint has invested more than 50% of its profits into this technology. And as predictive allocation technology improves, there could be up to a 90% reduction in miles travelled for same-day deliveries in London, according to the business. 

Elsewhere, it has made a shift to low or zero-emission vehicles for deliveries in urban areas. Each cargo bike that replaces a van cuts emissions by more than four tonnes of CO2 a year. By 2020, the company will have 100 bikes in Central London, saving more than 400 tonnes of CO2. The bikes are more nimble and 50% faster than a small van. There has been a 305% uplift in cargo bike job bookings since 2017. 

Meanwhile, CitySprint has doubled the number of its plug-in electric vehicles in the past two years, launched the first hydrogen van for logistics in the UK and its zero-emission fleet now makes up more than 20% of its total London fleet. 

This brings an annual saving of 38 tonnes of CO2e for 2019, based on current mileage rates. 

The information in this impact story has been supplied by CitySprint.

Region: