(01)
Description
Navigation system using Big Data applied to predict high demand areas for orders
Gracias al análisis de datos históricos en el consumo de pedidos de comida a domicilio, podemos predecir pautas en el consumidor y establecer zonas de alta demanda, para que el trabajador pueda tomar decisiones ágiles y aumentar su productividad.
(02)
Data Project
(03)
Process
Research & Synthesis
We must know and understand the delivery sector in a time of change driven by the Rider Law. To do this, we conducted a survey for both workers and consumers of home delivery. Light interviews with couriers and an in-depth interview with union leader Gus Gaviria were conducted.
There is currently a division between salaried delivery workers and those who remain as self-employed workers. Their claims are different, but the pain points are repeated in both groups.
Insights
Thanks to this process we were able to delimit the problem:
🎯 Working too many hours for low salaries. More than 53.3% work more than 40 hours per week. With the most frequent salaries being between 1,000 / 2,000€.
🎯 Rainy days are the most dangerous, but when there are more orders.
🎯 Oversizing of the sector. Due to the increase in workers assigned to the main platforms in recent years, the waiting time between orders has increased considerably.
🎯 Set the multiplier at low rates to be competitive. Lack of knowledge of how the algorithm works to get more orders.
(04)
Solution
Tool to humanize and streamline the delivery sector
The main functionalities of the application are defined as follows:
📍 Applied Big data to determine areas with higher demand
📍 Use of tagging to find interesting resources
📍 Live chat with connected couriers
📍 Forecasting calendar
📍 Display of incidents that occurred in the area
📍 Creation of routes avoiding incidents that occurred
Calendar
Show the peak hours of order history based on the selected day.
Navigation
Muestra las zonas de alta demanda con la relación entre el número de pedidos y trabajadores
Route
Creation of routes avoiding dangers and prioritizing the use of bike lanes.
(05)
