3 Steps to Achieving Predictive Logistics

A reporter recently asked me how, in my view, the shipping industry would make money in the coming years. The question caused me to reflect on how this might have been answered just ten or five years ago versus today. In the past, growth came from scale: mega-ships, port infrastructure, and buffer. Today, the answer is very different: shippers and service providers must operate smarter, not bigger. This is where profit will be made, and where the leaders will separate from laggards.

Defining Smarter Logistics

Smarter decisions require looking at the supply chain as a network of digital information opposed to physical assets. If harnessed correctly, supply chain data can be the key that unlocks new forms of efficiency, profitability and differentiated customer service. But achieving true data intelligence is no easy feat. Here are three essential steps to get there.

  1. Make Historical Data Machine Readable: Today’s supply chain data is dirty, siloed and disorganized. Its value is limited in this form. But by filling in blanks, sequencing and de-duping information, dirty historical data is transformed; it is now cleanly structured, making it machine-readable. Milestone events (like inland ETAs) as well as behaviors (such as cancellations) can now be predicted with high accuracy. A data intelligence platform with a proprietary data ingestion engine can unlock this value from historical data.
  2. Analyze the Past to Predict the Future: With clean data in-hand, machine learning can identify patterns in operational performance and customer behavior that are unrecognizable to the human eye. Simulation engines can calculate all “what-if” scenarios to determine the most probable outcome. But culling these insights is easier said than done because not all artificial intelligence (AI) is created equal. The shipping industry is highly complex and nuanced; improving predictions requires a platform custom built for the logistics industry. 
  3. Apply AI to Current Business Processes: Though seemingly futuristic, it’s best to think of AI simply as a different kind of tool to help solve problems– including core visibility and data challenges. This type of tool is available today and it can be leveraged with the data you already have in your hands. It does not require massive IT infrastructure overhaul nor investment, and it can fit into your existing business process and workflow.

We often ask our customers two questions: “If given the choice between an old flip phone and a smartphone, which would you choose?” This, of course, represents the ability to use AI over traditional methods. Secondly, “What if you could see into the future?” Picture having an accurate view of what will happen four or even eight weeks from now. Predictive logistics will enable shippers and service providers to move away from educated guessing to data-driven decisions, based on cleanly structured data and highly accurate predictions. This higher level of certainty will allow the shipper or service provider to be a better partner and drive unprecedented efficiency and profitability. This is the power of predictive logistics. 

Putting it Into Practice

Retailers and manufacturers deploy AI to, at the time of booking, accurately predict inland destination shipment arrival times to allocate volumes across air and ocean and meet inventory demands. A data-centric approach enables the retailer to better predict partner performance and container movement, drastically improving supply chain and, over time, reducing inventory on hand.

Similarly, a 3PL applies AI to predict with a higher-degree of confidence and accuracy the ETA for door delivery and inland location. Instead of relying on schedules to estimate future shipment events, predictions built on machine learning and simulation can be delivered directly to shippers. In this instance, the use of arrival time windows and confidence intervals provide accurate risk metrics around shipment delays and improved SLAs on behalf of their shippers.  

A terminal operator leverages AI to better predict container flows to improve its yard optimization, partnership to carriers, and differentiated service. With accurate predictions of container arrival by mode, operators can improve labor staffing, on-yard picks and positioning to increase throughput and performance for carriers, motor carriers, 3PLs and shippers.

A carrier who uses an AI-based data intelligence platform to ingest historical and real-time booking data can accurately predict booking fall-down to yield increased vessel utilization, customer service, and dynamic pricing capabilities. Highly accurate and granular container demand/supply forecasts can yield millions of dollars in potential profitability through reduced repo cost, container availability, increased velocity and contribution margin.

Key Takeaways: 

  • Look at your assets not as physical, but as digital assets — because the future of the supply chain is about data intelligence, not scale.
  • Leverage the data you have using a fundamentally different tool: AI. Use industry-tailored machine learning to cleanly structure and canonicalize that data, and use predictive logistics capabilities to make smarter decisions that differentiate your operations and service to customers. Be smarter, not bigger. 

Adam Compain is the CEO of ClearMetal, a predictive logistics company that uses data science and machine learning to unlock new efficiencies for global trade. Compain co-founded ClearMetal after working in Hong Kong at one of the world’s largest container-shipping companies. For five years prior, Adam deployed the newest geo-commerce technologies at Google, and for 16 years he has been the Executive Director of the nonprofit he founded to export charitable goods. Adam holds five technology patents, two degrees from the University of Michigan, and an MBA from Stanford University.