Trace Id is missing
September 16, 2024

Molslinjen develops an AI-powered dynamic pricing strategy with Azure Databricks.

Molslinjen is Denmark’s largest domestic ferry service and one of the world’s largest fast ferry companies. The company wanted to revamp its operations with dynamic pricing and demand prediction.
With the assistance of AI company Halfspace, Molslinjen created an AI analytics toolbox via Azure Databricks that provides data-driven innovation and modernized services.
The Azure Databricks solution has reduced fuel emissions, improved customer satisfaction, and brought in millions of additional revenue. It also won the 2024 Franz Edelman award for achievement in operations research.

Transporting across Danish land and waters

With over 15 million annual guests and a fleet of 21 ships, Molslinjen is Denmark’s largest domestic ferry service and, thanks to its catamaran ferries, one of the world’s largest fast ferry companies. In addition to its ten-route ferry service, the company operates over 116 bus routes in Denmark. When private equity firm EQT bought the company, Molslinjen found itself able to invest more in its technological development and aimed to bring the airline dynamic pricing model to its operations. This was both a huge opportunity to revolutionize Molslinjen’s business model and a huge innovation for the ferry industry. Molslinjen had already started building dynamic pricing solutions for its larger fleets about a decade ago, but it wanted to expand this pricing to other fleets as well as improve demand projection to better load the ferries.

Finding the right partner for a sustainable system

The dynamic pricing engine Molslinjen built was successful at securing the best price based on the guests who wanted to ride, while also giving less popular departures a price that encouraged more riders. But finetuning the models was a manual process. The team also built forecast engines to help optimize ferry loading, but didn’t have enough internal resources to scale up as quickly as the company wanted. They realized that having a solid cloud platform could accelerate their work even more. So, they looked for an implementation partner to help build it. “We needed someone to help us frame our data, develop it, and build upon the groundwork we had laid,” says Lasse Janerka, Director of Business Optimization & Digital at Molslinjen. Molslinjen turned to Halfspace, a multi-award-winning AI company that uses data-driven initiatives to help deliver better business outcomes.

Azure Databricks has a lot of features that easily process data and build models, and those tied together closely with the other Azure functions and resources we were using, enhancing our workflow efficiency and scalability.

Simon Cloetta Kristiansen, Co-Founder & Associated Director, Engineering, Halfspace

Ferrying across capacity considerations


On ferries, there isn’t really a magic number that determines vehicle capacity. “For an airline or a hotel, you have a specific number of seats or rooms you can sell. But for a ferry, you have space, and your capacity is how much you can take into the space,” says Mikkel Bjørn, Associate Director and Partner at Halfspace. Capacity changes depending on the size of the vehicles coming aboard and the placement of those vehicles, so Molslinjen needed that information in advance to optimize the space on board. “Cars come in many different sizes, from more than six meters in length, to less than three meters—at least in Europe. That makes revenue management a lot harder in ferry operations than in other sectors. In order to take that into account, we had to develop an advanced, customized capacity management module in our Databricks solution,” explains Bjørn. Since Molslinjen needed to know a fleet’s capacity in near real time, Halfspace needed to ensure the new system processed data as quickly as possible. When a car showed up at the harbor, it needed to show up on the application at the same time to ensure that the Molslinjen team could load as efficiently as possible.

Halfspace had the technological expertise to build the platform, while Molslinjen had the hands-on knowledge to make it work for the ferry industry. ”Halfspace has always gone out of their way to help Molslinjen,” says Janerka. “They’re opening our eyes to new technologies and ways of thinking. We teach them how to do things commercially. We understand each other.”

Halfspace considered a variety of Azure tools and decided that the crux of this advanced analytics platform would be Azure Databricks. “Azure Databricks has a lot of features that easily process data and build models, and those tied together closely with the other Azure functions and resources we were using, enhancing our workflow efficiency and scalability,” says Simon Cloetta Kristiansen, Co-Founder & Associated Director, Engineering at Halfspace. Azure Service Fabric is a crucial part of the platform, providing a centralized location for storing all organizational real-time data while enabling seamless integration with Azure services like Azure Databricks and Power BI This integration supports real-time data processing and visibility, essential for dynamic pricing models and operational optimization. Azure Databricks handles pipelines and updates machine learning models for dynamic pricing and demand prediction. Finally, Power BI connects directly to the SQL database to provide live data via mobile phones to employees loading the ferries. Halfspace also built a web-based application using Azure Static Web Apps.

 

Cars come in many different sizes, from more than six meters in length, to less than three meters—at least in Europe. That makes revenue management a lot harder in ferry operations than in other sectors. In order to take that into account, we had to develop an advanced, customized capacity management module in our Databricks solution.

Mikkel Bjørn, Associate Director and Partner, Halfspace

A scalable and cost-effective system


Molslinjen and Halfspace originally created this technology to forecast different departures in terms of the number of vehicles and type of vehicles. Now, Molslinjen uses the solution for a number of purposes, from planning how to load the ferries to using segmentation to drive marketing activities and improve catering. “What’s key is how we are leveraging these technologies to create a data-driven and AI-enabled organization,” says Janerka.

 

This powerful AI toolbox has brought in an additional 15-20 million DKK ($2.6-3.2M) per year. And it all started with just one route and Azure Databricks. “Azure Databricks was the foundation of this entire transformation,” says Cloetta Kristiansen. The earnings come from a combination of cost savings, increased revenue, and a 3% reduction in fuel consumption and emissions in 2023, which roughly totaled to 5000 tons of CO2. Molslinjen has also reduced delays which optimizes fuel usage since the ferry doesn’t have to increase speed and make up for lost time, and it is also more strategic about not overloading one section of a ferry with heavier vehicles, which creates drag and increases fuel consumption and emissions. “That was a welcome, positive side effect of the whole project,” says Janerka. “Better forecasting meant better loading of ferries for fewer emissions.”

 

Sailing to a greener future


Molslinjen has a vision of every route being green, carbon neutral, and electric. It believes it can achieve this with the advantages of capacity optimization and AI tools for price setting and commercial operations. This means better forecasting, better planning, better marketing, better catering, better staffing, and overall, better ferry operations.

 

As an important signifier of Molslinjen’s bright future, Molslinjen and Halfspace were awarded the 2024 Franz Edelman Award for Achievement in Advanced Analytics, Operations Research and Management Science. "The feedback for the win was that this is not only a transformation story about a ferry operator significantly changing how it operates its vessels, but also an outline for the ferry operator industry and the global shipping industry on how to forecast for lower fuel emissions,” says Nielsen. Molslinjen and Halfspace have successfully created an Azure Databricks advanced analytics toolbox that has not only transformed its operations, but the entire ferry industry.

 

Find out more about Molslinjen on Facebook and LinkedIn.

Halfspace has always gone out of their way to help Molslinjen. They’re opening our eyes to new technologies and ways of thinking. We teach them how to do things commercially. We understand each other.

Lasse Janerka, Director of Business Optimization & Digital, Molslinjen

Take the next step

Fuel innovation with Microsoft

Talk to an expert about custom solutions

Let us help you create customized solutions and achieve your unique business goals.

Drive results with proven solutions

Achieve more with the products and solutions that helped our customers reach their goals.

Follow Microsoft