Having a crystal ball is not a stretch!
The combination of artificial intelligence, machine learning and timely data can give us a window into the future.
Supply chain managers and partners are faced with the challenge of not just creating proactive processes to minimize the impact of disruptions, but to know when the disruptions might occur. This is the basis for ensuring the resilience and sustainability of the supply chain.
The first formal definition of resilience was introduced in 1973. It focused on the ability of a system to absorb unusual disturbances and remain functional. In 2013, the White House defined resilience as “the ability to prepare for and adapt to changing conditions and withstand and recover rapidly from disruption,” adding that “resilience includes the ability to withstand and recover from deliberate attacks, accidents, or naturally occurring threats or incidents.”
Resilience research has emphasized two important features: the inclusion of the post-event recovery phase and the use of functionality—at the component and system levels—as the primary framework for analysis. Most resilience metrics used in research relate to the “functionality recovery.”
One of the biggest changes to the business landscape over the last 20 years or so has been the significant increase in the level of risk confronting supply chains. We have moved from a world of relative stability – and hence predictability – to an environment which is characterized by turbulence and uncertainty with a consequently heightened potential for disruption to business activity. Supply chains are increasingly impacted by external events which can affect the availability of materials and their movement through the chains.
A national example of increased attention to supply chain resiliency is in Canada. Canada is devoting resources to secure supply chains that in turn ensures economic prosperity and national security.
More resilient supply chains are secure and diverse — facilitating greater domestic production, a range of supply, built-in redundancies, adequate stockpiles, safe and secure digital networks, and a world-class manufacturing base and workforce. Moreover, close cooperation on resilient supply chains among partners who share data through visibility platforms will foster collective economic and digital security and strengthen the capacity to respond to supply chains disruptions, environmental and other emergencies.
There can be no doubt that in recent years the business environment has become more turbulent and hence less predictable. Whereas in the past it was standard practice to plan – with a time horizon of months, if not years – now the challenge is to find ways to become much more responsive to events as they happen, or better to be able to predict an event in the future.
Data and information buried in moldering files has no value. But timely data that forms the foundation of operational and strategic planning forecasts – now that has value.
As supply chains are subject to almost continual stresses and disruptions, there is a growing need for every member of the complex supply webs to be able to anticipate changes and proactively create and implement policies and procedures to minimize the impacts.
Focused models based on machine learning artificial intelligence algorithms are rapidly emerging. Designing them to meet the needs of supply chain web operations and the constituent processes is growing in importance. Supply chain webs which were focused on just-in-time and more recently just-in-case, are being redesigned to merge the fundamental elements of both approaches based on the sometime competing characteristics of transparency, consistency, flexibility, resiliency, sustainability, and low cost.
While corporations have been at the leading edge of creating forecasting and alert systems, the complexity of the supply webs requires that all members need to contribute.
Airports and seaports play a major role in the movement of cargoes. A variety of functions include consolidating shipments, loading, and unloading air and ocean containers on and off aircraft and ships, inspection, transferring shipments between modes, etc. Addressing dwell times, determining available capacity – of trucks, rail, air, and ships – off-schedule transport services, and market changes (such as shifts in export markets from one country to another) are factors which terminal and port/airport authorities need to focus their attention; all related to system resiliency.
What are the key enablers of system resiliency?
Visibility and information sharing: The ability to see from one end of the pipeline to another is essential. It is important to be able to see the changes that are on the horizon both upstream and downstream. Information sharing provides a powerful platform on which to build collaborative working relationships across the supply chain.
Access to information on capacity: An important facilitator of flexible supply chain management is the ability to access information on capacity when required. Capacity here refers not only to operational capacity in transport and container handling like at marine terminals but also in storage as well. Furthermore, that capacity may not be owned by the firm in question, it could come from partners across the network and third-party providers.
Interoperability: In an ideal world, organizations would be able to alter the architecture of their physical supply chains in short time frames with minimal cost or disruption involved. Equally, those same companies need the ability to manage multiple supply chains serving specific market segments. To enable this reconfiguration, it greatly helps if the nodes and links of the supply chain are ‘inter—operable’.
Network orchestration: Because the achievement of higher levels of adaptability generally requires inputs from a variety of other entities in the wider supply/demand network, the need for co-ordination across the network arises. As supply chains become more ‘virtual’ than ‘vertical’ there is a growing requirement for orchestration/coordination.
How to address the need to view into the future
A recently completed project funded by Transport Canada at one of the nation’s largest airports demonstrated how with limited data, a tool predicted operational and capacity problems two to three months into the future. The model, drawn on air cargo capacity, shipment volumes, and destination markets, employed algorithms which identified patterns within the data to identify clusters of occurrences which coincided with events of disruption of air cargo flows.
An essential aspect of a machine-learning algorithm is the accumulation and analysis of continually new data. Cargo community systems provide valuable feedstock of information and data on which the algorithm can create and test findings which are translated into alerts for managers to assess, create proactive solutions and implement them as required. As demonstrated in the Canadian study the larger levels of timely data generated by these systems married to machine-learning based algorithms have the potential to provide operations managers and planners with a look into the future as to when disruptions can be expected to occur.
While experienced managers have built up their own resources and based on experience can reasonably estimate when a capacity/operational crunch might occur, the model takes away the role of the human in predicting an occurrence. The manager can then focus their expertise and knowledge in creating a recovery process which will minimize the impact of disruptions. Humans are recognized as having a superior ability to create responses to disruptions.
Constant challenges to norms stimulate innovation, the latest developments show that major logistics companies accept that and are investing. The challenges to the traditional air cargo models are enormous and those that don’t change will quickly become irrelevant. The companies that refuse to see that risk will not be capable of meeting expectations of BCOs for anything but low revenue non time specific air cargo. The benefit to the BCOs of proactive programs to address disruptions is to rely on routings and partners which meet their business model of sustainability.
While the initial model developed in Canada is not yet fool proof, by combining this initial phase of development with an avalanche of data from a cargo community system, the crystal ball will become clearer and more accurate.
About the author
Charles H.W. Edwards, B.A., M.Sc., MBA, has over 50 years in the transportation, distribution, and logistics industry. Edwards is a vice president of SASI World and a professor of the practice at the University of North Carolina at Chapel Hill in the Department of City and Regional Planning. He is a Scholar Fellow of Sigma Chi Mu Tau (Supply Chain) honor society. He began his career as a truck driver in Toronto. Since then, Edwards has worked in international freight forwarding in Canada and the UAE, numerous sectors of the airline industry, aviation design and manufacturing in Germany and the United States, ocean freight, rail management, economic development, and logistics education. Edwards can be contacted at charles.edwards@sasiworld.com.