Developing a Logistics Digital Twin for a Parcel Distribution Network in Mexico

Developing a Logistics Digital Twin for a Parcel Distribution Network in Mexico

Problem

Estafeta is one of the top three parcel moving companies in Mexico, with 130 facilities. They move between six and eight million parcels per month using a fleet of aircraft, trucks, and last-mile vehicles.

They wanted to understand how to best use their budgeted capital expenditure to adapt to the growing demands on their service of delivering parcels all around the country, as well as increase network throughput.

Solution

Estafeta hired VP Consulting, a Mexican consulting firm that works with value networks and applies digital twins for logistical challenges. They developed a logistics digital twin for Estafeta to analyze what was happening in their parcel distribution network, increase network throughput, and perform bottleneck analysis.

This digital twin was created using AnyLogic and included the software's features such as a built-in database, agent-based modeling, discrete-event simulation, and GIS mapping.

The built-in database was used to connect the digital twin to the real world using historical and minute-by-minute, parcel-by-parcel forecasted demand to see customer behavior. Agent-based modeling helped to organize the schedules and interact with the planes, trucks, and hubs. With discrete-event simulation, the company modeled every hub when considering resources and capabilities during the sorting and delivery processes. Finally, GIS mapping allowed the developers to scale the model and simulate transport behavior.

Using the digital twin model, VP Consulting was able to see and analyze the complex dynamic systems and logistics of the distribution network.

In the model, the UI allows all trucks and aircraft to be displayed, and even the exact time that every aircraft lands can be seen.

UI of the various distribution hubs throughout Mexico used for delivering of parcels

The user interface of the logistics digital twin showing the various distribution hubs

There are various graphs where different outputs can be seen. Examples of these are total demand and individual customer demand, both of which can be displayed at any time in the model. This means that decisions can be made promptly about customers, products, and services.

At every distribution hub in the model, statistics on peaks, valleys, and operations can be seen and analyzed to determine if there are enough resources available. Sorting and demand can also be observed and evaluated.

So, all the graphs are visible and give insight while the simulation is running. They can then be consolidated into an Excel spreadsheet to be analyzed afterwards to draw conclusions.

Results

The consultants found, using the logistics digital twin, that in the actual operation, two processes were overlapping, but they didn’t understand how this was affecting the rest of the operation. By using the available graphs in the digital twin, where the analysis can be done minute by minute, they could identify the exact moment the overlapping occurred and how it was affecting the operations for more than one hour. They could then develop a solution to this problem by reducing the bottleneck.

This was one breakthrough that helped increase network throughput by 21%. In addition, backlogs were identified, and it was understood, using historical data, that some backlogs were a part of life while others needed attention.

Through a combination of better synchronization, collecting, sorting, connecting, and delivering operations, capacity constraints across the network were identified and overcome.

In the beginning, Estafeta wanted to know how to better use their capital investment for growth. In the end, through a logistical digital twin, they understood that they didn’t need to invest any money but rather sequence processes at a specific location and synchronize their operations better. So, they delayed their $25 million investment for two years but continued to grow, maintaining profitability.

This logistics digital twin is currently being used for specific peak seasons to determine the best distribution model to implement. It is also utilized for some big clients to simulate the best way to achieve the goals of fulfilling the service agreements that Estafeta has with them.

The case study was presented by Claudia Vazquez del Mercado and Alejandro Nila, of VP Consulting, and Jesus Garcia, of Estafeta, at the AnyLogic Conference 2022.

The slides are available as a PDF.



Études de cas similaires

Plus d’études de cas

Obtenez une brochure avec des études de cas de l’industrie (en anglais)

Télécharger