As the world’s leading fireworks provider, French pyrotechnics company Etienne Lacroix ships fireworks globally and has supported events such as the Burj Khalifa Grand Opening Fireworks Show in Dubai and Bastille Day, France’s national holiday that is celebrated on July 14 each year.
Logistics play a major role for the global player in the pyrotechnics sector, as shipping hazardous materials such as fireworks requires special precautions, especially when crossing international borders. Labels typically include warning pictograms and complex regulatory information in various formats. Recipients such as customs officers must be able to tell at a glance what sort of goods are inside, what the required shipping conditions are, and who is authorized to open the freight.
“On top of that, customers have their own specific requirements,” said Eric Marini, director of Information Systems at Etienne Lacroix. “There are different color codes in place for hazmat, depending on the country of destination. Red may mean danger in European countries, but in China, for example, it stands for celebration. Green has a different meaning in the Middle East that it does in the U.S., and so on.”
With so many different regulations in place around the globe, labeling the shipments at the end of the production line is not only time-consuming, but also leaves a lot of potential for human error. Now, with support from SAP partner STMS and SAP, the prototype for a generative AI solution was introduced that has the potential to reduce human involvement to a minimum.
Business AI Co-Innovation Project
In early 2024, Sébastien Faure, general manager at STMS SOLUTIONS, a longtime partner of both SAP and Etienne Lacroix, participated in a Hack2Build event where SAP partners leveraged SAP technology to address pain points of their customers. The idea for a generative AI use case came up, and STMS approached its customer Etienne Lacroix.
At that time, IT and business experts at Etienne Lacroix were very interested in how generative AI could assist them in making their processes easier.
“But I have to admit, I was also a little skeptical,” Marini says. “Everybody is currently talking about AI, about how to introduce it to the industry and the huge benefits we will reap from leveraging it. But suggestions are scarce when it comes to implementing it in a way that guarantees the company will benefit from it.”
“Together, we looked at the pain points of the company and discussed possible use cases with SAP Co-Innovation Lab,” said Faure. “We received the requirement from the business that AI should help avoid human tasks that are error-prone and don’t actually add value.”
Supply Chain Use Case with Generative AI
The team quickly identified that the administrative process of creating labels for shipping was a task currently performed by human employees that required a huge amount of time and focus. The huge potential for the usage of generative AI was evident.
“With guidance from experts from the SAP partner organization, we were able to build a specific SAP app on SAP Business Technology Platform (SAP BTP) and SAP S/4HANA to organize and use AI,” Faure says. “We then collected feedback from the business experts at Etienne Lacroix and, in the next step, brought in the generative aspect.”
To meet the regulations from the customer, the developers from STMS aggregated all the necessary data and created a model on SAP HANA Cloud vector engine.
“Technically, this prototype leverages everything SAP has to offer in terms of AI right now,” said Faure.
The remaining human task is to validate whether everything on the label is correct. “That was the most important requirement,” Marini says. “Human intervention must be guaranteed, as with all AI use cases.”
“It’s the generative aspect that makes all the difference,” said Miliau Pape from SAP Co-Innovation Lab. The LLM suggests what should go on the specific label—such as warning pictograms—based on legal regulations, historical customer requirements, cultural standards, and so on.
The Retrieval-Augmented Generation (RAG) provides the LLM with context and makes the outcome relevant and reliable.
“Simply put, when the AI is trying to be as creative as possible, the RAG provides guardrails, so it doesn’t potentially go wild and come up with nonsense,” said Pape.
For each shipment, a prompt draws on destination, shipping route, specific customer data, such as the storage and language this customer required the last time, or the colors or signs used to indicate that the shipment contains hazardous materials and can only be opened by experts with a certain certification.
“This generative AI use case, at this point, may still be only a prototype, but it’s an applicable idea, an actual use case for an actual pain point in our company,” said Marini. “It is a very decisive first step in our AI journey.”
Read the full article here
