Presentation video: https://youtu.be/1zSJxyDMyCw
Marcelo Rodrigues dos Santos (Loggi)
Day: October 8th (friday), 10:30h
Loggi Supply – Last-mile Supply Forecasting
Abstract:
Loggi’s mission is to connect Brazil, delivering anything to anyone as quickly as possible.
Through technology (mobile, AI, automation, IoT), Loggi has created a next-generation logistics network and is, in an unprecedented way, positioned to unleash the growth of a new trade-in Brazil with a fast, cost-effective and reliable logistics solution.
Its current network is composed of tens of thousands of partner drivers connecting customers to hundreds of small distribution centers distributed all across Brazil, responsible for local operations. We also have a couple of large cross-docking facilities connecting the small hubs through a large scale network of ground and air transportation, responsible for national operations.
The mentioned distribution centers, named as agencies, are part of the last-mile activity.
Depending on the total of deliveries demand, a city can have several agencies. The last-mile delivery itself is carried out through the mentioned partner drivers, who are self-employed drivers (independent contractors), a fact that implies some specific challenges for efficiently allocating drivers to delivery routes.
It is known that driver habits related to hour of the day (time slot), days leading up to and following holidays, rainy days, strong traffic days, etc. could affect the driver’s motivation to accept a new itinerary. Analyzing these data could make it feasible to understand how much demand an agency is able to meet in a specific day hour and during the day.
Currently, Loggi’s agencies know the demand of last-mile deliveries for a given day, but have a reduced visibility on the supply (flow) of drivers during the hours of that day – what impacts its capacity to allocate drivers to itineraries. In this sense, the main business question is what is the hourly potential of dispatch of an agency? The visibility of the hourly potential of dispatch would allow agencies to best distribute and schedule the process for allocating drivers during the hours of the day. Besides, the hourly potential of dispatch information could help Loggi to:
● Plan the receipt of bags by the agency and its dispatch times
● Think about the agency’s work schedule and headcount
● Take a snapshot of the agency’s performance
To learn more about Loggi, visit: loggi.com/venha
Marcelo Rodrigues dos Santos is a Mathematician and holds a PhD in Information Sciences (UFMG). Currently, he is a Data Analytics Manager at Loggi.