Hey there, it’s Selim, Senior Director of Software Engineering at Careem.
I’m thrilled to share a paper developed from Careem’s latest academic collaboration, with Lahore University of Management Sciences (LUMS), one of Pakistan’s leading private research universities.
Our team of digital mobility specialists teamed up with the LUMs research team to introduce STEF-DHNet, a demand prediction model that combines convolutional neural network (CNN) and long short-term memory (LSTM) to integrate external features as spatiotemporal information and capture their influence on ride-hailing demand.
This model aims to address the challenges associated with forecasting demand in a particular region which is influenced by several external factors, such as time of day, weather conditions, and location. Accurately predicting demand for ride-hailing services is critical to developing more effective surge peak pricing strategies, improving driver positioning, and enhancing customer service.
While grid-based deep learning approaches have proven effective in predicting regional taxi demand, these models have limitations in integrating external factors in their spatiotemporal complexity and maintaining high accuracy over extended time horizons without continuous retraining.
We’re delighted to offer our colleagues an accelerated learning journey by empowering them to work on previously unsolved regional challenges and by partnering with regional academic leaders to study solutions to increase the economic development of our region.
I’m proud to say that our partnership with LUMS not only helped Careem increase the welfare of our captains and customers but also will inspire talent from our region to join us on our journey to simplify and improve lives through the mobility of people, things and money. Part of Careem’s mission is to incubate native talent and provide opportunities to address regional problems in their footprint. And this partnership with LUMS was a great step towards achieving this purposeful objective.
Once selected, the home grown talent will join our special programme ‘Careem Next Gen’ and become experts in their fields through a deep collaboration with our most talented engineering leaders as well as continuous customised technical training.
As we move forward, I express trust for many more regional industry-academia partnerships to bring the best of tech talent trained by our visionary leaders from Academia into Careem while helping us reach new heights in tech innovation. By partnering with academic thought leaders we can make a meaningful difference in our economy and continue our journey of becoming a lasting tech institution.