Digital technologies have changed the competitive dynamics of the logistics services industry. Companies like Maersk have started upgrading their traditional services towards technology-supported transportation solutions.
To offer a better customer experience with smarter, faster, and more sustainable logistics, companies must increase operational efficiency by addressing industry problems such as highly fragmented markets, low transparency, underutilized assets, costly manual processes, and, in many instances, outdated customer interfaces.
This new market environment requires the logistics companies to make it a strategic priority to invest in technologies for efficient performance, communication, and data visibility.
Technological advances have made it possible to incorporate tools in businesses that were previously impossible to manage digitally. Specifically, for the logistics industry, artificial intelligence (AI) and machine learning (ML) techniques are increasingly adopted to solve different business problems. Nowadays, with the challenge of container shortage and insufficient supply of shipping space in the industry, there is an impending need to improve the container turnaround and optimize the container transportation.
How can a digital platform solution match millions of import containers and reuse them for export in China? One attractive solution for most of the participants in this process is container triangulation: moving containers from importers directly to exporters without the need to empty containers to port. There have been different approaches to optimizing empty container movements using container triangulations (or street-turns), mainly deploying mixed-integer linear programming (MILP) and continuous programming. This research project investigated the current process and challenges of automation and digitalization of container triangulation in a digital platform developed by Maersk in China.
Automating and accelerating a leaner and greener solution
We divided the project into three principal stages: input, model, and output. At the input stage, we met with experts from the sponsor company as well as practitioners from markets spanning the globe. With a good understanding of the problem and experts’ insights, we framed and collected sample data from the sponsor company. Moving to the model stage, we applied match and machine learning clustering algorithms, and then we developed a MILP model to run the transport flow optimization. Finally, at the output stage, we quantified financial savings and environmental benefits based on the container distance traveled. Finally, by combining quantitative analysis and expert advice, we arrived at some recommendations.
What is next?
Container triangulation is a reality on the market today, as many companies, countries, and researchers are evaluating this new alternative. As reviewed above, container triangulation leads to reductions in transportation costs, lead time, and CO? emissions. By offering this process within its digital platform, Maersk can attract and retain key partners due to the incentives it generates. In addition, this process can be automated—as we observed during the project—planning and executing more quickly.
As we discovered promising benefits, we also found that there is still much to be done to fully realize the potential of container triangulation. Here are our takeaways and recommendations to scale up the solution into the digital platform:
• Container triangulation provides efficiencies in time, cost, and carbon emissions.
• Maersk can transform customer service by implementing digital solutions.
• Scale-up is about collaboration and coordination among stakeholders. It is essential, since their interaction is what makes the solution an attractive alternative.
For future studies, we recommend taking the concepts covered in this research to a global optimization perspective, considering different transport companies or partners. As more participants get involved, more data and more markets become available in the platform, becoming more attractive for users and creating a feedback loop. This may generate business growth.
Every year, approximately 80 students in the MIT Center for Transportation & Logistics’s (MIT CTL) Master of Supply Chain Management (SCM) program complete approximately 45 one-year research projects.
These students are early-career business professionals from multiple countries, with two to 10 years of experience in the industry. Most of the research projects are chosen, sponsored by, and carried out in collaboration with multinational corporations. Joint teams that include MIT SCM students and MIT CTL faculty work on real-world problems. In this series, they summarize a selection of the latest SCM research.