Metadata Moves: Knowledge Graph Technology for Logistics

The amount of metadata being exchanged during the movement of goods across the world is mind boggling. Just imagine the trail of metadata your evening glass of wine left across platforms and systems before the bottle arrived at your place for you to open and enjoy its aroma.

Just as the nose of the wine – be it floral, citrus, fruity or any other – can “speak” about its grape variety, the fermentation process, the aging and storage conditions, so too the metadata of its packaging can carry information about its manufacturing, storage, delivery, handling, packaging and other processes.

Think about it! Before the bottle reached your glass, it was part of a batch, produced by a certain winemaker, then packed into a case and stored at a certain temperature, ready to be shipped. Further, the wine cases traveled through transportation hubs, many of them crossed borders, changed vehicles and then finally got to a warehouse from where another round of logistics processes began through which wine cases reached retail stores, restaurants or you directly as a consumer.

Under the hood of all these supply chain processes lies metadata. And so does a great potential for eliminating supply chain waste and increasing the efficiency of logistics processes through better metadata exchange across systems.

The Metadata Weaving the Logistics Networks

From container tracking through loading onto a ship and further traveling through borders and arriving at a warehouse point, the set of events and the data they are logged by across the ecosystem of logistics processes, is large and heterogeneous.

There’s metadata at the level of the transport means – trucks, trains, ships, etc. Metadata at the level of the hubs they pass or wait at – terminals, warehouses, stations, borders, etc. There’s also a metadata trail accompanying these physical locations and activities at the level of documentation – documents about compliance, origin of goods and other things related to the actors who are part of the processes of manufacturing, transportation and selling.

Getting back to our wine example, here are some of the events (with their data points) the bottle in question undertook before getting to you. Depending on the region you got your wine from, it was probably first shipped into a container, as part of a case. Then that container was picked from a depo, sealed, brought to a gate, inspected for temperature and other requirements or shipping regulations. Next, it got loaded onto a vehicle, which may have been checked for an overload event later on the way or detected to drive over the permissible speed.

So, while on its journey, your bottle changed not only carriers and transportation hubs but also, in a way, its identity – from being  part of a batch to becoming a part of a shipping order. And all the way through this journey, it was accompanied by documents related to customs clearance, a certificate of origin, shipping instructions, a cargo certificate, etc. Given this complex dynamics of data exchange, amounting to all sorts of metadata, the success of logistics operations is closely related to the ability to trace and connect that data.

(All of these events were created following a sample view of shipping events in these slides: Towards a Shared European Logistics Intelligent Information Space.)

How A Knowledge Graph Built of Semantic Data Enables Smooth Logistics Operations

There are plenty of standards governing and providing guidance on the best way to create a meaningful knowledge graph that can ensure the interoperability of all the moving parts in the logistical process. We will explore those in a next post about the importance of supply chain traceability supported by EPCIS 2.0. It’s the standard providing a Web compatible universal language for supply chains – a critical building block enabling data interoperability for supply chain transparency.

Now, let’s telescope to the bigger picture of how a knowledge graph built the semantic data modeling way can help for smooth logistics and transport operations – namely, the potential to look at all this metadata from a 360-degree view. And when it comes to managing data from heterogeneous sources and offering interoperable solutions, knowledge graphs have proven to be the answer across various verticals.

Providing the basis for interlinking multiple sources through data integration, a knowledge graph can serve as a backbone technology for real-time monitoring of logistics processes and as a platform for exchanging key data across systems. And when it’s built with semantic technology, it can also allow smarter storing, managing and retrieval of data. Specifically for logistics and transport data, semantics can enable smoother data integration across systems as well as more efficient data exchange throughout supply chain events and operations, thus eliminating drivers of waste in the logistics industry, among which errors in picking, sub-optimal routing, document re-keying and errors, inefficient networks.

The Benefits Of A Knowledge Graph For Logistics At A Glance

Logistics, by its nature, connects vast amounts of events, activities and actors. In the same way, knowledge graphs interlink descriptions of concepts, entities, relationships and events. Click To Tweet But while the complexity of the logistics processes make for difficult logistics scenarios, the interconnectedness of the data related to these scenarios opens vast horizons for efficient problem-solving.

At a glance, some of the business benefits of data integration and management through knowledge graphs built with semantic technology for logistics data are:

• optimal loading of vehicles;
• space utilization in warehouses;
• reduced time in transit;
• smoother flow of information between actors (manufacturers, shippers, consumers, etc.);
• greater supply chain transparency.

Towards Smoother Flows And Faster Logistics And Transport Services

In their book The Logistics and Supply Chain Innovation Handbook Disruptive Technologies and New Business Models, authors John Manners Bell and Ken Lyon point to a number of drivers of waste in the logistics industry, among which:

• capacity underutilization;
• sub-optimal routing and driver behavior;
• document re-keying and errors;
• inefficient networks;
• paper documentation;
• rekeying of data.

Following these waste drivers are innovation drivers that could eliminate that waste such as data standardization, automation, IoT sensor technology, open data, dynamic routing systems, to mention just a few.

Seen through the perspective of semantic technology and its capacity to meet data and knowledge management challenges, the above-listed inefficiencies (and the potential for their elimination) can be solved through a semantic layer on top of the metadata in the systems.

Adding semantics to these traveling data can address issues resulting from data silos and incoherence. With semantic technology, the level of cooperation through data exchange can be improved and thus the infrastructures involved in the logistics and transport processes can be supported and enhanced.

Epilogue: Cheers To The Benefits Of Knowledge Graphs For Logistics

To get back to our glass of wine, a knowledge graph (i.e., an integrated architecture with connected metadata from various sources) allows efficient, safe and transparent logistics processes. It “connects the dots” of your wine trails.

Integrating all the data from the storage vessels, the transportation vehicles and the hubs the wine has been in allows for the optimal usage of the infrastructure as well as easy regulation compliance.” When this data is managed expertly this enables timely access to it and a high level of cooperation among the agents involved in all the logistics and transport processes.

And if you are like us at Ontotext, you will enjoy your wine even more.

Here’s to the benefits of data integration & management through knowledge graphs for logistics!

Interested in exploring how your logistics business can benefit from knowledge graph technology?