Artificial intelligence (AI) in the supply chain is a chicken and egg problem. Some have praised AI for its potential to increase visibility into supply chain operations. In other words, AI first, visibility second.
This may have been true when widespread, real-time supply chain visibility was not otherwise possible. But to transform supply chain AI, including extremely powerful generative AI that generates new insights, outcomes, processes, and efficiencies from large datasets, we need to flip the equation. First comes visibility, followed by GenAI-driven innovation across the supply chain.
A local retail manager, distributor, manufacturer, or procurement person can wake up on Monday, fire up a familiar AI chatbot (and in some cases even voice-activated) and see if their supply chain is optimized for the week. Imagine asking someone in natural language: If not, ask how you can adjust your supply chain to meet your goals. GenAI enables this interaction with supply chain systems.
But the only way a GenAI-based supply chain solution can automatically provide such answers is by knowing the status, location, condition, movement, etc. of every product, box, case, pallet, etc. in the supply chain. This is the case. And the only way to know is if the product itself can communicate information automatically without human intervention. Today, this is possible through a ubiquitous visibility platform called the Ambient Internet of Things (IoT).
GenAI in the supply chain
Estimates from global consulting firm Ernst & Young 40% of supply chain companies are investing in GenAI. They use his GenAI to map complex supply networks, run “what-if” scenarios, predict upstream and downstream supplies, and use chatbots to help partners get answers more easily. and even generate new contracts based on past or existing contracts.
In these cases, companies are training their own AI models, historical data, and whatever they can collect from partners. Next, they ask his GenAI to find ways to increase efficiency. However, as EY analysts note, “GenAI tools are only as powerful as their input data, so they are limited by the quality and availability of data from supply chain partners.”
But the ultimate goal of supply chain AI is to generate new routes, processes, product designs, and supplier lists based on real-time data and execute them as quickly as possible (even faster than humans can).Or as a manager said to harvard business review“When a supply chain crisis occurs, the key to staying competitive is finding alternative suppliers before others do, because everyone is trying to do the same thing.”
This requires training GenAI solutions on more and more up-to-date data about real-world supply chain operations. Ambient IoT is here.
Ambient IoT: The language of the supply chain
Ambient IoT digitally signs products, packaging, and locations. Digital signatures are a real-time supply chain visualization language that will ultimately be incorporated into the Large-Scale Language Model (LLM) that is the basis of GenAI. These signatures are transmitted via IoT pixels, self-powered, postage-stamp-sized electronic tags attached to anything in the supply chain that needs to be tracked and monitored. The IoT Pixel contains its own computing power, sensors, and Bluetooth communications, allowing him to describe the journey of products and packaging through the supply chain in terms of data that his LLM can consume. Ultimately, they will create a bridge between the physical and digital worlds, making supply chain data available for the first time that can actually view, predict, and optimize operations.
Ambient IoT pixels are easily deployed, off-the-shelf, standardized bridges or gateways installed through an established mesh of existing wireless devices such as smartphones or wireless access points, or installed in stores, warehouses, delivery trucks, etc. communicate data through. Indeed, with proper permissions and privacy protections, ambient IoT pixels can extend supply chain visibility to consumers, convey data about product use, reuse, and recycling, and serve as the basis for more advanced GenAI models. can be proven.
And it sends data continuously. Unlike the supply chain records used to train GenAI models today, ambient IoT data describes the supply chain. right now. With this visibility, all you need to do is implement GenAI to answer the question, “What am I seeing in my supply chain?” right now? ”
Real-time visibility and ambient IoT data generation across the supply chain could also help address one of GenAI’s challenges. That said, the data used to train LLMs will always reflect unintended data biases from the sources of generation, which often include a company’s various ERP systems.
Products tracked through the supply chain using ambient IoT speak an objective truth. Because products actually show up where ambient IoT says they exist. Ambient IoT also minimizes human error by not requiring workers with RFID scanners to track deliveries.
Ambient IoT data accurately describes the route and time a product takes within a supply chain. Products also include data about the parties and facilities involved in their handling in a digital product passport. If applicable, ambient IoT pixels can add information about temperature, humidity, and carbon emissions at every stage to the LLM.
According to EY, one area where supply chain companies are considering using GenAI is regulatory and ESG reporting. The best and most cost-effective way to collect vast amounts of data so that GenAI can generate compliant information is through ambient IoT.
From chatbots to automation
On a daily basis, there are two ways in which the convergence of ambient IoT and GenAI can benefit supply chains. First, more people in the supply chain will be able to understand changing conditions and take proactive steps to optimize or modify supply chain operations. You don’t need to be a data analyst or procurement expert to ask GenAI chatbots about shipping status or contact alternative suppliers, but LLM and GenAI tools can evolve to produce useful results. Companies will continue to need data experts to ensure that But democratizing supply chain analysis and research could enable the rapid decision-making needed to improve competitiveness.
Second, GenAI and other AI tools can help build a bridge to further enhance supply chain automation. Through machine learning, particularly reinforcement learning commonly found in control systems, software can be trained to make decisions that achieve better outcomes. For example, they could eventually be trained to proactively detect supply chain disruptions and automatically dispatch alternative suppliers or shippers. Or you can initiate predictive maintenance by determining whether a particular warehouse or manufacturing system or line is at risk of failure.
This is done by learning from large datasets including supply chain data generated by ambient IoT.
As we have learned in recent years, complex supply chains exist like razor blades. Several small factors can lead to confusion. Artificial intelligence is essential to avoid future disruptions. But to get there, supply chains will need to unlock data that they currently cannot see. Ambient IoT provides visibility data that will be the foundation of tomorrow’s GenAI innovations.