Of all the latest digital technologies slated to disrupt supply chain operations imminently, artificial intelligence (AI), along with robotics (which arguably, you might classify as an AI application), is the one making the most tangible current impact on the industry.
“It is difficult to think of a major industry that AI will not transform. This includes healthcare, education, transportation, retail, communications, and agriculture. There are surprisingly clear paths for AI to make a big difference in all of these industries.”
—Andrew Ng, Computer Scientist
Perhaps your company is among the half of all enterprises already deploying AI to solve supply chain problems or improve performance.
Whether it is or not, though, you might find the following examples of AI application in supply chain operations interesting, intriguing—or even inspirational.
AI in transportation management
Statistic: Companies that use TMD applications have achieved freight cost savings of up to 8%.
Source: ARC Advisory Group Strategic Report.
The ROI of transportation management solutions (TMS) has been proven many times over. However, with the integration of AI technology, new platforms are taking TMS capabilities to unprecedented heights, offering even more potential for companies to gain efficiency in transportation and reduce freight costs.
Even without AI onboard, TMS applications can save money and help operators to increase service performance. They can do this by:
- Helping resource planners to optimise routes for truckload and less-than-truckload shipments
- Identifying when and where multi-stop routes prove more economical than single-stops
- Highlighting the comparative performance of carriers
- Analyzing data to answering questions such as “Which specific geographic areas are impacted most often by late deliveries?”
TMS that learns
When the TMS tech-stack includes a layer of artificial intelligence, shippers and 3PLs can gather and use an even broader array of data about transportation performance.
For example, they can use live weather and traffic congestion data to create more accurate, dynamic route plans, and real-time vehicle tracking can provide them with comparisons of planned versus actual route performance.
Better still, machine learning enables all of these functions to be carried out without first building complex models, since the closed feedback loop (planned routes versus actual) enables the AI-equipped TMS to train itself by perpetually comparing inputs with outcomes.
Integrated AI in warehouse management systems
“What’s actually enabling a lot of the warehouse automation, when you look under the covers of it, is the advancement being made in terms of artificial intelligence, machine learning, deep neural networks —these sorts of capabilities,”
—Thomas O’Connor, Gartner Analyst.
Just as the success of TMS is enhanced by integrating artificial intelligence, so too is that of the warehouse management system (WMS). Several WMS vendors are integrating machine learning into their platforms to create new opportunities for companies to improve productivity and efficiency in their warehouses and distribution centres.
Nowhere is the value of AI in WMS more evident, perhaps, than in the e-commerce fulfillment centre, where it is often necessary to introduce customers’ orders, dynamically, into the picking workflow.
In these high-velocity environments, there is no time to gather orders into batches and release them to the warehouse in waves, so many eCommerce companies have shifted to waveless picking, a process for which artificial intelligence is eminently suited.
WMS for the e-commerce age
Machine learning technology is increasingly being used to drive warehouse management and control systems, since it can parse data from multiple sources to calculate the time required to complete warehouse tasks. It can then use the results of those calculations, which factor in rapidly changing priorities and conditions, to optimise the way assets and resources work.
Such systems can interpret inputs and adapt to changes more quickly than any human could, which is why they hold so much value for e-commerce operations. AI is the one viable solution for balancing pick density targets with multiple SLAs and dynamic introduction of single orders, the frequency of which is highly unpredictable.
AI as an aid to procurement communications
The potential uses of artificial intelligence in supply chain management are many and varied. For example, aside from enhancing the power and capability of systems like WMS and TMS, AI can enable computers to perform tasks traditionally reserved for humans.
Computer vision, natural language comprehension, and similar sensory developments have all been made possible by developments in AI technology, essentially giving machines the ability to see, hear, and communicate in similar ways to people.
One global beverage leader recently drew on AI-powered language comprehension and human-machine interaction to improve the speed and effectiveness of communication between itself and its suppliers.
“Chatbots are important because you won’t feel stupid asking important questions. Sometimes talking to someone can be a bit intimidating. Talking to a chatbot makes that a lot easier!”
—Petter Bae Brandtzaeg, Associate Professor, University of Oslo.
The company implemented a chatbot, integrated with multiple systems, to solve issues relating to the inefficiencies of human-to-human dialogue in its procurement operations.
The issues included:
- The need for the beverage company’s employees to call supplier helpdesks to obtain information.
- The time wasted while helpdesk operators accessed multiple systems to accumulate and then share the requested details.
- The labour costs involved in running operations 24/7 to maintain productivity.
Making helpdesks more helpful
The company used the chatbot’s language processing and conversational interface to provide text-based interactions between procurement employees and suppliers’ systems and between the suppliers’ employees and its internal business information solution (SAP).
Authorised users could open the chat interface to perform queries, to which artificial intelligence provided the answers in moments, due to its ability to interrogate multiple systems, answer questions in text conversations, and pull up relevant files and documents for scrutiny.
The solution cut the cost of executing and resolving routine queries, allowed the participating companies to reduce support hours, and freed helpdesk agents to focus attention on more complex tasks for which their skills added value to relationships between the company and its suppliers.
Other uses and use-cases for AI in the supply chain
Statistic: Around 32% of supply chain professionals are actively using robotics and automation.
Source: MHI/Deloitte, 2019 Annual Industry Report.
In addition to the three examples we’ve shared above, we’ve been hearing about companies using artificial intelligence to solve a wide range of supply chain problems. The following examples include some you will undoubtedly have come across (like robots and AGVs), and a few that you might not have considered.
- Optimization of truck and trailer loading to maximise space utilization
- Improving the accuracy and efficiency of pick-to-light warehouse systems
- Compression of order cycles with the use of optimized fulfillment-source allocation
- Enhancing spatial-awareness and other capabilities of warehouse robots, using machine learning and computer vision
Whatever position you take in the perpetual debate surrounding business automation and its impact on future jobs, the success of your company could soon come to depend upon artificial intelligence.
Slowly but surely, it’s rising as a pervasive game-changer in enterprise generally. With supply chain superiority being a crucial factor separating the best from the rest, though, the procurement and logistics arenas may well be the next proving grounds for ever-more-sophisticated AI-related technologies.