Optimizing Logistic Fleets with Machine Learning to Enable Sustainable On-Demand Deliveries in Cities

Project Information

Airmee

Grant agreement ID: 827075

DOI

10.3030/827075

Project closed

EC signature date 31 July 2018

Start date 1 August 2018

End date 30 November 2018

Funded under

• INDUSTRIAL LEADERSHIP - Innovation In SMEs

Total cost

€ 71 429,00

EU contribution

€ 50 000,00

Coordinated by

AIRMIZ AB
Sweden

Periodic Reporting for period 1 - Airmee (Optimizing Logistic Fleets with Machine Learning to Enable Sustainable On-Demand Deliveries in Cities)

Reporting period: 2018-08-01 to 2018-11-30

Summary of the context and overall objectives of the project

Urbanization, GDP per capita, and the rise of e-commerce are all underlying factors that not only drive online sales and the last mile delivery market but also the increase in urban logistics in general.

As a growing number of vehicles in urban areas implies increased congestion, air pollution and
noise, which negatively impact traffic safety, urban economic competitiveness and quality of life in general, more and more cities are experiencing issues related to last mile delivery of goods. Many cities have started to understand these challenges, and attempts are being made to solve these problems. However, they are mainly related to passenger mobility but very little is done regarding the last mile delivery of goods. Last mile delivery of goods is a more difficult issue due to its complexity and many stakeholders.

With the rise of e-commerce, consumer preferences have grown increasingly important in the formerly business-oriented parcel-delivery market. Large e-commerce players and logistics providers have identified last-mile services as a key differentiator. Consumers demand transparency, speed and convenience in the deliveries and these demands have become major decision-making criteria when choosing where to shop. E-commerce players are therefore working hard to offer the best delivery experience, especially by improving the speed and accuracy of delivery times.

The large and traditional logistics players have however a hard time keeping up with these changes as these customer-driven deliveries put pressure on existing processes, technologies and infrastructures. Adding to the challenges for traditional logistics players are higher fuel costs, government regulations regarding environmental concerns and the lack of transportation management systems and tracking systems for last-mile deliveries.

The logistics industry is therefore expected to undergo major changes both in terms of the makeup of the market and processes. Startups that can act faster are likely to continue to gain market share in the growing last mile delivery market. Technology will also be a key factor for logistics providers if they are to reduce costs, increase efficiency and provide better services that are in line with customer preferences.

To solve the issue of meeting customer needs while ensuring environmental sustainable logistics, the objective of the projects was to study the requirements for a technological logistics platform for last mile deliveries and the business model for it.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

Today, traffic dispatching, requires significant manpower and any automation of this process will significantly improve efficiency and reduce costs. The main challenge that these companies have is that there isn’t any software that can handle the often complex operations a logistics provider in an automated way.

During the project we have approached and researched the operations of several different type of logistics providers. The main challenge that these companies have is that there isn’t any software that can handle the often complex operations a logistics provider in an automated way. The existing software solutions that are being used have all been developed by software companies with limited expertise and knowledge in logistics operations. The main need we found among logistics providers was the ability to automate much of the operations that is mainly traffic dispatching, which require significant manpower. The ability to automate dispatching, which includes allocation of delivery assignments, scheduling, and routing, was also the feature that was lacking in the various software platforms that logistics providers were using. The optimization of logistics fleet is highly complex, an NP-hard problem, and still an unsolved problem in research.
We found that several parameters have to be taken into account in the optimization algorithm such as type of vehicle, the speed of which each courier delivers, type of item, and changes in the route.

We also received great interest in our platform and services by several of Sweden’s largest online retailers. By providing them with our delivery services the retailers have been able to provide their customers in turn fast and convenient deliveries. This has enabled to gain data and iterate on the platform faster.

Our platform is in an ok shape. It is implemented with modern and mostly up-to-date technology. It is of small size and besides the algorithms of average complexity. While there are no critical issues in the assessed projects, there are a number of improvements needed, in order to reduce accumulated technical depth and ensure future maintainability. In particular, the test coverage is low and should be improved. Yet, due to the small size and modularity the effort should be reasonable.

Our current platform would require numerous months of work to replicate with the right competence. For each day as we gain more data and add more complexity to the platform it will require more resources to create anything that is of equal or similar value.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

Our platform is in an ok shape. It is implemented with modern and mostly up-to-date technology. It is of small size and besides the algorithms of average complexity. While there are no critical issues in the assessed projects, there are a number of improvements needed, in order to reduce accumulated technical depth and ensure future maintainability. In particular, the test coverage is low and should be improved. Yet, due to the small size and modularity the effort should be reasonable.

Our current platform would require numerous months of work to replicate with the right competence. For each day as we gain more data and add more complexity to the platform it will require more resources to create anything that is of equal or similar value.

Our world-leading technology has allowed us to test our platform with customers to reduce the number of vehicles on the road by optimzing the logistics fleet more efficiently than today's methods.

Through the project we were able to focus on a) deeper research into features and functions needed b) automation, optimizing and improving the allocation, scheduling and delivery algorithms, which is most important, and c) on improvements (efficiency, flexibility, transparency and analytics) in the platform.

Airmee logo