Inbound Logistics | July 2023

Validate the information. Advancements are quickly rolling out to improve large language models (LLMs), but users should be cautious of using these systems for important decisions without domain expertise validation. –Alex Schwarm, Head of Data, Arrive Logistics Assess the potential costs associated with

implementing AI, including technology acquisition, infrastructure setup, data preparation, talent acquisition or upskilling, and ongoing maintenance. Start small and experiment and collaboratively develop a roadmap for AI adoption that aligns with the organization's strategic goals, resources, and capabilities. –Kary Zate, Senior Director, Marketing Communications, Locus

social science specialists from Halmstad University and Gothenburg University in Sweden collaborated for more than three years to develop and trial an AI-based semi-autonomous voyage planning system. The project showed how AI and machine learning can enable more energy-efcient voyage planning for ship operators. The results demonstrated successful energy efciency optimization based on estimated time of arrival.

In addition, AI can reduce product return rates by analyzing customer data and making personalized product recommendations. Chatbots and virtual assistants can also help ecommerce customers through the returns process, taking on a large volume of customer inquiries and allowing human workers to focus on higher- value tasks. Chatbots can learn from customer interactions, honing their responses to improve the efciency of returns processes. Lastly, machine algorithms could help companies determine the most efcient and cost-effective way to handle returns, taking into account brick-and-mortar locations, warehouses, shipping routes, and carrier performance.

code and provide tiered classication data based on suggestion algorithms. Users can enter specic product descrip- tions to improve accuracy; meanwhile, machine learning models continuously improve the algorithms. STREAMLINING CUSTOMS CLEARANCE. Aside from helping com- panies ll out customs paperwork, AI solutions can also stream- line customs clearance processes. For example, KlearNow.AI, a pro- vider of customs clearance and drayage software, launched an articial intel- ligence-powered platform, Customs Engine, to enable companies to move goods efciently and compliantly through customs and ports. The new software platform, targeted at importers, customs brokers, and forward- ers, leverages ve years of deep learning from its AI and machine learning data ingestion tool which created data sets for submission to customs authorities. Customs Engine’s automated data extraction and document digitization capabilities can eliminate manual data entries and related errors. It generates Importer Security Filings and transmits them to customs authorities.

INCREASING LASTMILE EFFICIENCIES. “AI can help solve the famed last-mile problem

with smart sensors on delivery vehicles, manual driver input, or location-based tracking,” says Hehman from TXI. Its ability to process data and reduce human error can translate into huge opportunities for efciency improve- ments in the last-mile space. “Since moving products from a regional hub to a point of usage is usually the most expensive and challenging part of a supply chain, AI-enabled gains go a long way,” Hehman says. “By analyzing weather data, trafc patterns, and other environmental factors, an AI system can use data from smart sensors to ensure materials are delivered properly while optimizing the experience for the delivery person.”

SUPPORTING SUSTAINABILITY INITIATIVES. AI-powered tools can

cleanse and integrate data from dispa- rate sources to facilitate carbon emission measurement and reporting. One AI-enabled solution is BlueNode, which measures carbon and Scope 3 emissions from ports, terminal operators, maritime and rail carriers, shippers, and trade authorities. Everstream Analytics recently acquired BlueNode to expand its inter- modal analytics solutions and let users make data-based decisions on mari- time carbon mitigation—balancing costs, shipping time, and environmen- tal impact. AI-based solutions can also enable more energy-efcient sea voyages, a recent trial found. Maritime technology company Yara Marine Technologies, AI application developer Molow, and Chalmers University of Technology and

HELPING ENSURE WORKER SAFETY. By creating optimal routes, increasing

efciencies, and taking care of mundane tasks such as customs paperwork, AI can help boost worker satisfaction. It can also ag potentially dangerous situations for human workers. “AI can supplement human decision- making to ensure organizations are ensuring the health and safety of their employees,” says Hehman. “For example, a system can automatically send a message recalling delivery drivers when temperatures reach extreme highs or when environmental factors indicate tornadoes could form in the area.” n

PROCESSING ECOMMERCE RETURNS. AI can attack reverse logistics challenges from

multiple fronts. Using machine learning algorithms, companies can glean insights from their returns data and identify pat- terns and underlying causes. Retailers can then make the adjust- ments needed—from revising product information on their website to tweaking packaging to changing shipping meth- ods—that could lower the return rate.

146 Inbound Logistics • July 2023

Powered by