By Kasturi Datta |
Munich Fire Department in Germany worked with Microsoft to develop an AI operator that handles non-emergency dispatch calls using natural language.
The tool was built in Microsoft Foundry and several tools were used in the design process, including Azure Speech to ensure that the system supports human-like dialogue.
The operator ensures non-emergency patient calls receive the assistance they need while emergency patients are attended to by dispatch staff. This helps to reduce work overload.
“If patient calls come in alongside emergency calls, the emergency calls are prioritised and the patient calls must wait, and that’s the biggest hurdle,” said Julia Voss, dispatcher at Munich Fire Department.
With Foundry Azure AI search, the operator validates addresses against a municipal database and reduces the waiting time for transport from the hospital to be arranged.
“When a patient presses the emergency button, the nurse must run there immediately-but sometimes she’s on the phone with us ordering a patient transport. She then puts us on hold, which slows down everything,” said Florian Dax, member of the Munich Fire Department’s technology team and one of the chief architects of the operator.
While the operator can collate relevant information and aid nurses and healthcare workers, Microsoft Foundry’s governance system will ensure it remains within the boundaries of its assigned task, and the human dispatcher is the one making any executive decisions.
“We came up with a saying, ‘Der Bot hilft, der Mensch rettet’, it means the bot helps but the human saves,” said Mathais Duensing, head of IT-architecture for Munich Fire Department. “If the chatbot doesn’t understand something, or the nurse gets frustrated, they are connected to a human dispatcher immediately.”
The operator is scheduled to move into real-life testing later this year in the emergency department at LMU Klinikum, Munich’s biggest hospital.
“It’s important for us to be part of shaping AI in emergency services. We’re a niche-we don’t have thousands of branches- but in our world, AI can make emergency response significantly safer,” said Dax.