How the Virtual Watch Tower Complements Private Data with Public Data

[By Mikael Lind, Wolfgang Lehmacher, Xiao Feng Yin, Xiuju Fu, Kenneth Lind, Zhou Rong, Bart Coppelmans, Torbjörn Rydbergh]

It is often claimed that big data analytics and artificial intelligence improve our factual base for supply chain and logistics decision-making. Analytics is about what is, what may be, and what are my options if I wish to influence the flow of goods. This requires that private data stored in various systems of different actors is leveraged and shared in part with others along the chain, but also complemented with public data provided by different sources of the supply chain ecosystem. Upstream actors know at first hand when incidents happen which is valuable information for actors downstream; and information about weather conditions can help to predict the speed at which cargo moves. The analysis on alternative options, e.g., new routes, also benefits from access to publicly available traffic information or incidents reports. The more data is available the higher the chance for high quality decisions.

The shipper-driven and terminal-centric virtual watch tower / VWT (www.virtualwatchtower.org) or VWT network / VWTnet applies a novel approach and concept to data sharing and a foundation for extended supply chain visibility and collaboration. Through data sharing, collaboration, and co-creation the VWT fills a gap complementing existing freight management systems. The VWT enables actors along the chain to improve decision-making in supply chain and logistics management through the use of public and shared private data for descriptive, predictive, and prescriptive analytics. The VWT members share insights by utilizing pre-agreed private supply chain datasets across the community. These private datasets include plans, e.g., estimated times of arrival, and progress of cargo moving through the global end-to-end (e2e) supply chains. Private data benefits from being complemented by public data as public data provides additional insights into actual and future conditions of the supply chain ecosystem that can positively or negatively impact the flow of goods and cause early or late arrivals, which both are divergences from the original transport plan. An example is tailwind that can shorten flight times. Adding public data to private data improves the accuracy of the outputs of predictive and prescriptive analytics.

The public data that can for example be added is AIS data on ship movements, various kind of relevant traffic data, weather data, inland waterways water levels, blank sailings as well as flight and freight train cancellations, truck attributes and restrictions, locations of environmental zones, rules and regulations like driver rest times, and hazardous materials regulations. Public data can provide planning and performance critical insights, e.g., technical regulations as well as driver rest times. Omitting to take such data into account can impact transport reliability. E.g., omitting the mandatory overnight stays during route planning has implications for the timely execution of a transport. Furthermore, information on labor disputes, strikes, or high absenteeism in a port or terminal which may cause productivity losses and potentially congestion, or ad hoc changes in customs regulation, or simply a traffic jam or storm doesn’t appear early-on on the radar of cargo-owners and transport operators.

The VWT complements private data with public data to establish a more holistic view allowing for better planning and steering of operations. This is a perquisite for transport reliability and optimal use of resources, in particular in intermodal transport systems and a must for just-in-time production networks.

In this article, we focus on public data and provide examples mainly taken from maritime supply chains, as they are central to global trade and commerce. We also describe how public data is obtained and leveraged by the VWT and how relationships between public data/analytical service providers and the VWT are structured.

Definition of public data

Public data refers to information and datasets that are made available to the general public, stakeholders, or interested parties for access, use, and dissemination. This data is typically sourced from government agencies, regulatory bodies, research institutions, and other public sources. Public data in the supply chain and logistics context can include a wide range of information related to different modes of transports, ports, airports, rail, road, and river/inland waterways infrastructure, safety, weather and environmental conditions, trade and consumer trends, and much more. In maritime supply chain networks the central component are the sea-bound transports, including feeder services.

Public data (in supply chain and logistics) falls into two main categories.

1. Free Public Data: This refers to open domain data that is provided to the public without any cost. It is freely accessible and can be obtained without the need for subscriptions or payments. Examples of free public data in the supply chain and logistics industry include weather and sea conditions reports, road, rail, river/inland waterways, airport, and seaport traffic statistics, weather information, environmental monitoring data, publicly available regulatory information, and (limited content of) tracking data (e.g., AIS). Public data also includes information of roadworks and rail infrastructure maintenance as well as parking spots for trucks.

2. Non-Free/Payable Publicly Accessible Data: This category includes data that is available to the public but may come with certain access restrictions or require payment for access. While the data is publicly accessible, it may be subject to licensing agreements or usage restrictions. Examples of non-free/payable publicly accessible data in the supply chain and logistics industry include specialized research reports, premium weather forecasting services, truck telematics data, commercial high-resolution cleansed live and historical AIS data captured with AIS terrestrial, satellite, and vessel antennas, vessel particulars, some types of shipping indices, and certain detailed vessel databases.

Borders between the two categories are blurred and datasets can evolve from free to non-free and even from public to private. Multiple public data sources are freely available or accessible, but consuming data in an aggregated and processed way will make it a private dataset. An example is road traffic data. A consumer can access traffic data through consumer applications like Google and HERE, but if a company wants to use this data in an enterprise application a license is required. In the following, we are not revisiting these intricacies and focus on public data in general.

Public data plays a vital role in enhancing transparency, supporting research, analysis and evidence-based decision-making, and fostering innovation. It empowers a broad range of stakeholders in the e2e supply chain ecosystem, such as carriers, brokers, forwarders, governments and government agencies, like port authorities, customs and inland security and national disaster response agencies, research institutes, industry associations, like the International Federation of Freight Forwarders Associations (FIATA) and International Air Transport Associations (IATA), and trade association like the World Trade Organization (WTO) and International Trade Organization (ITO)  to optimize activities and improve their outputs.

In the context of the VWT, public datasets need to cover entire e2e supply chain networks and all modes of transports. A few examples of such public datasets are:

• Schedules of train / flight / ship etc. movements
• Data on cargo movements of the different modes of transport
• AIS data on ship movements (position, speed, heading, draft, destination etc.)
• RFID, camera and GPS data on train passages
• Sea condition data (e.g., tides and current)
• Government regulations and changes of the regulations relevant to all modes of transport
• Risks and disruption events along supply chain
• Information on location, equipment etc. in intermodal terminals
• Data on trade regulations, customs procedures, and import/export requirements

Also, data from fields not directly related to supply chain and logistics like trade flows, market trends, consumer behavior etc., informing for example forecasts, supply chain planning, are part of the list.

Categories of public data

In the VWT context, public data refers to data that is accessible to everyone and not being conceived as private data. Public data is an important complement to private data to augment supply chain and logistics visibility and intelligence and increase the accuracy of analytical outputs, such as the prediction of arrival times and supply chain disruptions. Public data may inform actors that a high number of ships are approaching a port to be served by limited resources possibly causing congestion and consequently delays of cargo, or that a road has been closed, or a train derailed, or that the levels of water of a critical inland waterway has plummet, as recently happened with European river Rhine.

The VWT can leverage a wide variety of public datasets, sources, and providers to improve the accuracy of its analytics (see examples in Table 1).

Data Category	Description	Exemplary Public Data Providers
Transport schedule	Schedule of train/shipping liners	Freight train schedule (e.g., traingeek.ca), Rail Freight Corridors (RFC), and data provided by Rail Network Europe (RNE) and national transport agencies, Sailing schedule providers (e.g., https://www.linescape.com/)
Weather data	Historical, current and forecasted weather data	National Meteorological Agencies (e.g., NOAA, Met Office), private weather data providers (e.g., Weather.com)
Navigation and Routing	Chart data and other information	National Hydrographic Offices, e.g., NOAA (USA), UKHO (UK) International Federation of Surveyors and their national members
	Tide and current information	National Oceanic and Atmospheric Administration (NOAA), National Tidal and Sea Level Facility (UK)
	Real-time vessel tracking data	Automatic Identification System (AIS) data providers, e.g., S&P Global, Orbcomm, Sprire, MarineTraffic, VesselFinder, FleetMon
Vessel Information	Capacity, type of cargo, design speed, length, breadth, draft, age, safety and cargo handling features, propulsion, fuel types and capacities, environmental performances etc.	Maritime Authorities, Class Societies, private companies like S&P Global, Clarksons, and Gibsons
Port Operations	Port infrastructure information	Port Authorities, Port Management Companies
	Berthing availability	Port Authorities, Port Schedulers
	Port opening hours	Port Authorities
	Vessel traffic data	AIS data providers, Port Authorities
Environmental Monitoring	Sea temperature	National Oceanic and Atmospheric Administration (NOAA), European Environment Agency (EEA)
	Water quality	Environmental Protection Agencies, Research Institutes
	Marine wildlife observations	Marine Conservation Organizations, Research Institutions
Risk management	Maritime accident and incident data	Marine Accident Investigation Authorities, e.g., NTSB, MAIB
	Maritime security threat data	Maritime Security Agencies, e.g., US Coast Guard, European Border and Coast Guard Agency (Frontex)
	Maritime regulations and requirements	International Maritime Organization (IMO), National Maritime Authorities
	Historical incidents and accident statistics	Maritime Insurance Companies, Reinsurance Companies, commercial data provider
Research and Development	Maritime technology and innovation data	Research Institutions, Industry Associations, commercial data provider
	Vessel design data	Naval Architecture Firms, Shipbuilding Companies
	Maritime logistics data	Logistics Companies, Shipping Companies
Government regulations	Country commercial guide and import/export regulations	Government relevant websites (e.g., trade.gov)
Risks and disruptions	Disruptive events and potential risks to the supply chain	Global disruptive event providers (e.g., https://www.everstream.ai/)

Table 1: Illustrative examples of public data and public data sources

VWT use cases require public data

What public data is required to enrich private datasets stored in the systems of different actors along the chain depends on the specific use cases that are addressed. The use cases range from enhancing supply chain visibility and re-planning, to improving risk and sustainability management. Over time, the VWT community might decide to address the following use cases which can all benefit from public data feeds:

1. Supply chain visibility and re-planning (foundational requirement): Access to public data, including tracking information such as AIS data, publicly available flight or train information, and road traffic data, enables cargo owners and other actors along the chain to complement private data to better monitor the movement of shipments as well as factors that may influence the flows of cargo, allowing for more accurate calculations of arrival times and if required proactive downstream replanning in case of delays.

2. Supply chain efficiency and optimized routing: Complementing private data with public data, like information on traffic situations and weather conditions, allows cargo owners and transport operators to make better informed decisions on the most cost-effective routes.

3. Risk/Disruption Management: Public data on weather forecasts, risk events at transport nodes, and safety incidents across supply chain networks allows cargo owners and transport operators to predict deviations from plans, disruptions and delays improving their risk management capability.

4. Customs and Compliance: Public data on trade regulations, customs procedures, and import/export requirements assists cargo owners to comply with laws and regulations. Compliance helps streamlining clearance processes and optimizing costs for all stakeholders involved.

5. Sustainability Management: Access to public data on greenhouse gas emissions of different means of transport, environmental impact of certain behaviors, and sustainable shipping practices enable cargo owners and transport operators to improve their ability to make well-informed eco-friendly choices, e.g., prioritizing those carriers and logistics service providers that apply eco-friendly practices.

The VWT aims at equipping actors in supply chain and logistics with actionable information and data reaching beyond the scope of traditional freight management systems and solutions for example to assist them in taking more accurate decisions, improving their risk management capability, while ensuring compliance and reducing Scope 3 GHG (greenhouse gas) emissions across supply chain networks.

Organizing public data flows

The VWT facilitates access to both raw and cleansed public data, as well as analytical services, for specific use cases, offered to individual VWT, and clusters of VWTs across VWTnet. Figure 1 illustrates flows of public data in the VWT.

Figure 1: Illustrative framework of public data flows in VWTnet

While the power of attorney facilitates the sharing of private data across VWTnet, the following enables the sharing of public data and access to analytical services.

• Each VWT member will be registered as a local VWT instance managed by the VWT; members include cargo owners, forwarders, carriers, terminal operators, and providers of data and analytical services.
• Both free/payable data and analytical services as well as their programming interfaces such as APIs and URIs will also be registered and managed by the VWT.
• The VWT manages the data and analytical services registry, access rights and authentication of the individual VWT instances, and the connections among VWT instances.
• VWTnet will facilitate the exposure and access to public data and analytical services provided by individual actors.
• The VWT may use web crawler to gather freely available public information, such as regulatory data for the purpose of informing, notifying, and alerting local VWTs.

There are different views on the importance of the harmonization of datasets. One school of thought believes that harmonization is necessary, another thinks that through the rapid developments of artificial intelligence (AI) harmonization will gradually lose its importance. In artificial intelligence (AI) standard setting bodies the term harmonization is currently being replaced by interoperability due to increased computational bias through “harmonization” in the literal sense of the word.

While the VWT is designed in a way that allows integrating necessary/desirable non-free/payable public datasets and analytical services, the contractual relationship with the data/analytical services provider is organized at user level and stays with the VWT community member that uses such data/analytical services. While they are valuable contributors to the co-creation of the VWT, the VWT will not engage in commercial arrangement with data and analytical services providers.

Especially in the open data realm actors should be aware of precedents where insurers didn’t cover damages related to incorrectly scrapped or web-crawled public data that was blended with proprietary information to inform decision-making. This risk is augmented by the use of unstructured-non-licensed data sets to inform AI-boosted recommendations. Mitigation of risk is related to using open data protocols, transparent charters for data use, including liability clauses further down the chain, etc.

Concluding remarks

Access to data is critical for businesses in the 21st century. This is particularly valid in the self-organizing ecosystem of supply chain and logistics. The VWT initiative adopts a distributed approach to data sharing and analytics empowering each VWT community member to combine and leverage own, 3rd party, and VWT datasets and analytical services. The VWT network follows the principal idea of a federated network of platforms connected to each other using the logic of shippers’ end-to-end supply chains. This is in line with the European Data Governance Act putting emphasis on “mechanisms to increase data availability and overcome technical obstacles to the reuse of data”. This enlarges the members’ view on supply chain networks and the cargo moving through them as well as their set of capabilities opening doors to new possibilities to deal with rising supply chain and logistics challenges. Private data sharing and access to public data and analytical services are the supply chain table stakes to reach required levels of visibility and transparency, and improve decision-making.

VWTnet contributes to the broader effort of digitalizing the supply chain and logistics industry. It is important to note that there is a symbiotic relationship between digitalization and collaboration. Neither can exist without the other, because they co-determine economic fitness. Successful partnerships co-evolve their collaboration through cooperative digitalization to contribute to an emerging era of digital symbiosis. Collaboration is a core component of the VWT concept. The cargo owner-driven and terminal-centric VWT responds to today’s operational needs by applying a lean system-of-systems approach shaped by its members. The growing community welcomes new applications from actors of the supply chain and logistics ecosystem.

Acknowledgement

We acknowledge and appreciate inputs received from Jan Bergstrand at Swedish Transport Administration, Martin Hullin at the Datasphere Initiative, Szymon Oscislowski at the European Commission, Mikael Renz at Swedish Maritime Administration, and Johan Ruthström at StrategyObject.

About the authors

Mikael Lind is world’s first (adjunct) Professor of Maritime Informatics engaged at Chalmers, and Research Institutes of Sweden (RISE). He is an expert contributor at World Economic Forum, Europe’s Digital Transport Logistic Forum (DTLF), and UN/CEFACT. He is co-editor of the first two books on maritime informatics, and is co-author of Practical Playbook for Maritime Decarbonisation.

Wolfgang Lehmacher is partner at Anchor Group and advisor at Topan AG. The former director at the World Economic Forum, and CEO Emeritus of GeoPost Intercontinental, is Advisory Board Member of The Logistics and Supply Chain Management Society, Ambassador F&L, Advisor GlobalSF, Advisor RISE, and member of the think tanks Logistikweisen and NEXST.

Xiao Feng Yin is a principal scientist of Institute of High Performance Computing, Agency for Science, Technology and Research (A*STAR) of Singapore. He has led various grants and industry projects in the areas of maritime study, logistics and supply chain. He received both his Master and PhD degrees from Nanyang Technological University (NTU), Singapore.

Xiuju Fu is Maritime AI Programme Director and senior principal scientist at Institute of High Performance Computing, Agency for Science Technology and Research (A*STAR), Singapore and active in developing and applying AI, big data intelligence, simulation, and optimization techniques for complex system management. Currently, she is leading Maritime AI Programme in Singapore for research in maritime data excellence, maritime AI modelling excellence, maritime AI computing and application excellence. 

Kenneth Lind is Senior Researcher at Research Institutes of Sweden (RISE) and has driven several research projects focusing on system architecture and software engineering challenges in the automotive and transport sector. He holds a PhD in software engineering from Chalmers University of Technology and has 20 years of industrial experience as technical leader.

Rong Zhou is a principal scientist of Institute of High Performance Computing, Agency for Science, Technology and Research (A*STAR) of Singapore. Her research interests include supply chain and logistics management, supply chain risk management, and production scheduling. She received her PhD degree from National University of Singapore (NUS).

Bart Coppelmans is leading the Global Industry Solutions team at HERE Technologies. A customer facing industry and solution consulting business development team responsible for creating business growth. He has a business economics background with over 15 years of leadership experience. Areas of expertise include business development, strategy consultancy and change management.  

Torbjörn Rydbergh is founder and Managing Director of Marine Benchmark and has a M. Sc in Naval Architecture from Chalmers and has been in the shipping and car industry for the last 25 years. He has worked for IHS Markit, Lloyd’s Register, and Volvo Cars among others.