Can AI predict the future of intermodal transportation?

Can AI predict the future of intermodal transportation?

In surveying the vibrant activity of the Free Port of Trieste it is easy to imagine how the past has prepared the present of this three hundred-year-old gem of intermodal transport. Yet Artificial Intelligence promises an even brighter future for supply chain logistics in proposing visions of Interconnected Cities, Smart Ports, Smart Highways, Logistic Marketplaces and Remote-Controlled Cargo Ships.

Plotting a course towards the future of intermodal transport requires understanding the current challenges facing intermodal transportation, accounting for the nature and the potential of AI, and applying a practical framework for transforming data into action. What does artificial intelligence promise, how do the particularities of the industry condition its potential applications, and where can public and private actors start building for tomorrow? In this introduction to our discussion in Trieste, let’s explore in turn the specificity of this market, separate AI facts from fiction, and propose our AI Roadmap as a practical roadmap for management in the months to come.

The Context

Intermodal transport involves a set of logistic challenges and opportunities concerning the transport of containers of goods using multiple modes of transportation. Intermodal transportation is more than just a collection of linear business processes: the potential of the logistics ecosystem grows exponentially with each additional mode of transportation. The ideal service is thus dependent upon the both the scope and the nature of the ecosystem – each transportation node offers alternative and often contradictory industrial logics. The value of the business hinges on each operator’s ability to propose an optimal mix of delivery times, flexibility, reliability, and information exchange.

If port environments constitute diverse hubs of the intermodal system, they have numerous characteristics that distinguish them from traditional corporations. Ports host intricate partner networks of administrations, terminals, shipping lines, trucking and logistics companies, and off-dock storage providers. They operate in offering a variety of private and public services with differing costs and resource constraints. Their potential is conditioned by their past – their activities and boundaries reflect the multiple business visions and interests that have interacted in their hinterland over the years. Like all complex adaptive systems, they have no single point of control, no universal set of objectives, no ideal state of equilibrium.

Regardless of their history, location, and resources, intermodal ports are faced with a set of common economic challenges today: intense competition, high fragmented manual processes, underutilized assets, legacy applications and formats and rudimentary customer interfaces. The coordination of container transport chains depends on efficiently meshing different processes and interfaces throughout the network. Improving intermodal transportation implies increasing the diversity and frequency of transportation options, providing faster terminal operations, and more fully utilizing port equipment.

In some cases, this will require simplifying the routines that characterize the different interfaces of the chain using clear international norms and standard container sizes. In others, this implies synchronizing opening hours in the chain, total optimizing route utilization, and balancing return loads. Meeting each of these challenges requires the availability of advance demand information, faster feedback from shipping companies to transport inquiries, and more ambitious information management between the port authorities and partner organizations.

The Opportunities

Artificial Intelligence (AI) involves the application of technology, data and analytics to replicate human thought and action. Machine Learning (ML), a subset of AI, relies on algorithms and statistical models to leverage data in improving either individual or collective decision-making. In practice, machine learning focuses on providing new knowledge about market challenges, whereas AI aims to mimic human intelligence in solving business problems. Although Machine Learning is often used as a synonym for AI, they imply very different visions, objectives, methods and applications in improving intermodal transportation.

In which areas of intermodal transport can artificial intelligence make a difference? In Infrastructure Management, smart sensors can help operators track, operate and maintain the physical facilities they manage. In Cargo Handling, monitoring systems can ensure that cargo-handling equipment operate efficiently and productively. Terminal Appointment Systems can help minimize turn times, reduce port traffic and contribute to better air quality. Finally, Customs and Collections Systems can improve the handling of cargo details and payments, including trade licenses, customs clearances.

In theory, machine intelligence can be leveraged to address four distinct types of challenges in intermodal transport. In the Back Office AI can pay dividends in assuming detail-oriented, repetitive tasks in accounting, finance, human resources, and legal counsel. Execution is yet another area of potential gains where AI and robotics can leverage computer vision systems, conversational interfaces, and autonomous vehicles to augment traditional workforces in sorting, inspection, retrieval, and delivery.

In Predictive Logistics Machine intelligence can be employed profitably to reduce the uncertainty and volatility that traditionally handicap key business processes like Network Management, Demand and Capacity Management, and Route Optimization. Finally, AI might well be leveraged to address the Production of Public Goods that stem from coordinating shared services, assuring service consistency, and providing environmental stewardship that elude the purview of private economic actors.

The Road Forward

If opportunities to leverage AI abound, the question of where to start provides a major challenge for port authorities and operators alike. It can be helpful to visualize the port environment as an economic Commons, or shared-resource system, producing goods and services for the benefit of current and future generations. Five elements characterize such Commons: the vision, the customers, the services, the technologies and the outcomes. Taken together, this shared resource system provides a bridge between the past and the future:

How does the business strategy reflect a vision of the intermodal port producing goods and services serving a common interest? As a shared resource system, Intermodal transportation networks produce both private and public goods. If the operators’ business models and processes are designed to produce and marketable goods and services, the ports themselves and the surrounding infrastructure are uniquely positioned to produce public goods for the benefit current and future generations. Route optimization, air quality, security, and general well-being are all examples of such goods that are by their nature non-excludable, non-rivalrous, and produce positive externalities. What specific services are your customers looking for and how do your data practices support the production of these services?

Which customers’ needs are you are catering to in this space: those of the general public, the shipping lines, the logistics companies, and/or the off-dock storage providers? In an AI Ready organization, tactical and operational information from your partner network will be collected, cleaned and validated, and then leveraged using AI to produce measurable benefits for your customers. Yet, the current scarcity of data documenting freight transportation issues continues to handicap efforts to leverage AI. Data on movements in port environments and the surrounding metropolitan areas often lack details of trip origins and destinations and the industry/commodity breakdown for products being carried. Shipping weight data for international exports (truck, rail, pipeline and mail) hampers efforts to analyze current network bottlenecks and forecast future transportation needs. How will your organization obtain the quantities and qualities of data needed to build a foundation for your digital services?

Which AI-based services can bring this vision to life? How do you see artificial intelligence’s role in supporting the future of your organization? Are the benefits of AI in your vision tied to automating business processes, interpreting consumer input, mastering conceptual relationships or eventually influencing environmental dynamics? Artificial Intelligence isn’t a universal cure for your business, depending on the role port authorities wish AI to play, distinct quantities and qualities of data, algorithms, use scenarios, and metrics are needed. Using the AI Continuum as a visual aid, we suggest that AI can be used to Act, Predict, Learn, or Create. Given your specific objectives and context, several use cases can then be analyzed, discussed, and actioned:

No alt text provided for this image How do information technologies support these services? Where can you benchmark your vision with current use cases on the market? Fleet Management Systems like Verizon Connect and Jobber help operators improve fleet efficiency in analyzing their operational data. Transportation Management Systems like Transplace and Transporeon permit the integration transportation-related activities and participants in the logistics process. Digital Marketplaces like TimoCom and Uber Freight improve market efficiencies by bringing together shippers and carriers in matching freight capacity to actual demand. Virtual Forwarders like Flexport and Saloodo! provide customers with end-to-end contracts as a one-stop shop for transportation- related services. Finally, Tender Platforms like Ticontract and Colo21 allow shippers and carriers negotiate longer-term transportation contracts while reducing the carriers’ network imbalances.

With what metrics will you measure the success of your efforts? Information technology doesn’t solve customer challenges, managers do. How will organizations leverage AI to transform the data into collective action? In our AI Roadmap, we analyze how your customers take operational decisions, and identify where AI can provide measurable business value to smart port scenarios. The Roadmap looks at the four cardinal points of managerial decision-making: perception, prediction, evaluation and action to explore how each point conditions our ability to execute, invent and innovate based on the data at hand. Using this roadmap, we can map out the challenges and opportunities of matching human and machine intelligence based on specific business context of intermodal transportation.

No alt text provided for this image We will be discussing and developing each of these propositions at the Trieste Intermodal Day Conference on September 19th. Together with the other keynote speakers, we will discuss ways of mapping out your future vision of the business. We will extend the discussion to explore how the interplay between human and artificial intelligence will add value to in intermodal transportation. We look forward to seeing you in Trieste.

Dr. Lee SCHLENKER September 05, 2019

Lee Schlenker is a Professor of Business Analytics and Community Management, and a Principal in the Business Analytics Institute His LinkedIn profile can be viewed at You can follow the BAI on Twitter at

Marketing Data Ethics: Identity, Trust and Value(s)

Marketing Data Ethics: Identity, Trust and Value(s)

On Teaching Data Ethics

On Teaching Data Ethics