Artificial intelligence has been around for decades, but recent advances are transforming the technological landscape — including supply chain, logistics and transportation-management applications. AI is being infused into a variety of supply chain technologies, and promises to have a profound impact on how goods are moved around the world in the coming years.

The industry is still undergoing sweeping digital transformation, from manual to technology-driven processes, so it should come as no surprise that adding AI to the mix can create complexities. Following are five major challenges that supply chain and logistics organizations face when adopting AI-powered tools, along with some suggested solutions.

Getting Started

This can prove difficult for some companies. It can be a daunting task to remodel business processes with AI. Start a small project to address some meaningful portion of a specific business process, to learn the capabilities and limitations of AI. In this way, you can start building up a comfort level with the technology.

Employee Resistance

AI can be deployed across numerous processes in the supply chain, impacting many team members, so employee acceptance is vital. Internal pushback, however, may be an impediment. For example, employees might be concerned that AI will automate their tasks and eliminate or fundamentally alter them. Others might simply be resistant to change, a common problem when implementing any new technology. 

To ease the transition to AI-augmented processes, communicate with employees about how it can make their work easier, empower them to do their jobs better and — perhaps most importantly — improve customer service. Explain how AI can provide greater visibility across the supply chain, more detailed partner vetting, and more accurate forecasting, to name just a few of many use cases. Also, communicate how AI can help provide a competitive advantage for the company as a whole, strengthening job security in the process. 

Effective training helps the team understand the importance of AI and how roles might change in a positive way. In addition, choose user-friendly applications that make it easy for employees to clearly see the advantages of AI-based tools or capabilities.

Trust

Trust issues can be a barrier to artificial intelligence. After all, AI isn’t infallible, and people can also make mistakes. Mistrust can be alleviated over time by showing proven results delivered by AI-supported applications within your organization.

On the flip side of trust, however, you need to take necessary precautions. Understand the risk factors — such as percentage of accuracy for a particular product — especially when allowing AI to make decisions or calculations autonomously. Human oversight is still critical at this stage in AI maturity, and it’s equally important to have professionals who are trained to work with the technology and understand what it can and cannot do.

Data Quality

Data drives supply chain and logistics technologies, but AI is only as good as the information you feed into it, so data quality is a major consideration. The ability of AI to analyze incredible volumes of data, sometimes in real time, and derive valuable insights on such key elements as delivery ETAs, partner performance, carrier qualifications and global trade trends. But if those insights are based on inaccurate information, AI output has little value and can even be detrimental.

One effective approach is to use a single, up-to-date data source of the truth rather than relying on multiple sources. In addition, companies must establish processes — even to the point of using specialized tooling — to ensure high standards of data cleansing, which involves removing duplicate, incorrect or irrelevant entries.

Industry data augmented by AI insights can help to uncover new market opportunities, find new suppliers and benchmark on-time delivery performance. But when engaging with a suitable vendor, be sure and collaborate with a trustworthy provider who’s committed to delivering comprehensive, up-to-date, quality data.

Data Access

Data silos can be a roadblock on the digital transformation journey, and that’s a particular challenge for AI. Traditionally, each department or group within an organization maintains its own data sets, but this paradigm is not a viable model in the AI era. Organizations must break down conventional silos and make valuable information accessible, in an appropriate standardized format, to AI-enabled systems. Data accessibility can make the difference between AI success or failure.

There’s no question that AI comes with some challenges, but the time is now to start thinking about how to overcome obstacles. AI tools will be incorporated into virtually every class of supply chain and logistics technology, and it will start to become table stakes — perhaps earlier than many expect. As AI continues to mature, and use cases develop, AI will become a critical competitive differentiator in supply chain and logistics, and those who lag in adoption may well be left behind. 

Ken Wood is executive vice president, product management at Descartes.