SPECIAL ADVERTISING SECTION
TECHNOLOGY : LOGISTICS VISIBILITY
Global businesses are embracing logistics visibility throughout their operations . International freight forwarders and assetbased carriers are taking advantage of up-to-the-minute tracking and monitoring of shipments and products . Worldwide brands are getting deep insight into what their customers are looking for . We ’ ve never had a better understanding of what ’ s going on throughout the supply chain than we do now .
But where do we go from here ? How can artificial intelligence and machine learning be utilized within the supply chain ? How do brands , carriers and other businesses level up their logistics visibility operations ?
It starts with better decision making based on a deep analysis of logistics data . Businesses can then use that analysis to accurately predict future demand , making it easier to meet customer needs and do business with supply chain partners . Breaking down those barriers also reduces waste and increases revenue .
Although logistics visibility helps us understand what has happened or is happening , it ’ s also critical to anticipate what ’ s coming next to determine subsequent steps . AI and machine learning can make a real difference at all stages — from validating and understanding the data we have to making accurate predictions about the future .
One of the most powerful aspects of logistics visibility is the ability it lends companies and carriers to predict and respond to significant events . The Russia-Ukraine and Israel-Hamas wars — combined with issues in the Red Sea and elsewhere — illustrate the need for companies to react quickly .
Taking that one step further is AI and its ability to predict when circumstances deviate from the norm . For example , in instances such as a snowstorm preventing flights from taking off in Denver , Colo ., a truck becoming stranded due to a mechanical delay in Seattle , Wash ., or a container ship blocking the Suez Canal , AI can rapidly analyze the disruption at hand , illustrate the impact and provide suggestions about how to respond .
At nVision Global , supply chain disruptions themselves aren ’ t the only consideration . Moreover , it ’ s about understanding these disruptions and combining that knowledge with clients ’ changing needs .
“ Our clients are looking for adaptability , both because of these changing political environments but also because of changing capacities and cargo types . They want adaptability in the software and services they use ,” said Stewart Dunsmore , senior vice president of supply chain services at nVision .
Cutting through complexity is vital for adaptable supply chains . Logistics data can be very noisy due to disruptions along with larger shifts in supply and demand . Cleansing and validating the data is essential to understanding what ’ s going on and provides a solid foundation for predictability and forecasting . This is one area where AI can really shine .
“ When you go through a process , you ’ ve got a certain amount of predictability ,” Dunsmore said . “ As we deploy AI , it monitors that predictability , and then it adds more functionality to its monitoring .”
The key here is that AI is selfcorrecting . As it starts to identify trends , patterns and outliers in the data , AI can identify where there might be errors and adjust for them . It then learns those errors and the corrective processes for each one . As AI increasingly analyzes more logistics visibility data , its machine learning algorithms help it become even more accurate .
“ As you continue to create these blocks of corrections , you reach a certain point where you have a highquality process because any errors are automatically being corrected by the AI ,” Dunsmore added .
An improvement of 1 % to 2 % in data accuracy may not matter much for smaller visibility datasets , but when these incremental improvements are scaled up across national or worldwide processes with millions or billions of data points , they have an enormous impact on costs , efficiency and lead times .
With these potential results , it ’ s easy to think that AI is a solution for every problem . However , while it is an incredibly powerful tool , there is reason for caution . Expecting AI to improve everything simultaneously limits its usefulness . Data overload is a real issue , and a carte blanche approach means it ’ s easy to miss opportunities in the sheer amount of AI outputs . Using a focused method is preferred , as deploying AI on specific logistics processes allows for deeper understanding and makes it easier to act on the data the algorithms are presenting . nVision takes a compartmentalized approach to using AI on its platform by gathering a full understanding of a specific logistics process , then using AI to glean the right data and drive improvements . This is one way importers can use AI to improve the accuracy of their customs invoices .
“ One example is making corrections to Brazilian customs invoices , so that you never have a penalty fee coming from the Brazilian government ,” Dunsmore said . “ When the Brazilian government sees an error on a customs invoice , they quarantine the freight and then send penalty fees to the end customer , making it an issue between the supplier and the customer . AI can take care of that .”
Dealing with problems at the source helps companies speed up the receipt of goods and improve supplier and customer relationships . But how can these separate and discrete processes be combined across the whole supply chain ?
AI is a logistics game changer when predicting supply , demand and capacity . This is especially important across global operations , as logistics managers must balance local marketplace demand with international sourcing , manufacturing and shipping constraints . It ’ s here that AI-driven forecasting can help brands compete domestically while taking advantage of global economies of scale . nVision ’ s data gathering , forecasting and AI are used by an international brand to optimize supply chains and meet the needs of local markets in Europe . nVision ’ s software begins by gathering sales data from European retail operations , which provides an overview of the demand in local , regional and national marketplaces .
Next , demand is analyzed and replenishment and fulfillment numbers for each marketplace are created . This aligns supply and demand with information about where consumers are buying products . Machine learning algorithms then combine tracking and product flows across the supply chain , including distribution center inventory levels , manufacturing times , transit durations , original supplier outputs and other resources . This allows the AI to predict availability from multiple inputs .
The end result is an integrated supply chain that ensures products are available when and where needed across Europe .
But what if consumer tastes change ? Perhaps a restaurant brand is expanding into a new market and wants www . joc . com September 9 , 2024 | Journal of Commerce 55