Innovate to enhance the quality of our services and maintain a high level of customer satisfaction.
We have integrated artificial intelligence into our customer support process, thanks to the expertise and work of Meryem Cherif, Industrial Engineering, Artificial Intelligence and Data Science Engineer, specialising in Machine Learning, Deep Learning, Natural Language Processing (NLP) and data analysis. The model used is capable of analysing time series using advanced statistical methods and Machine Learning, offering considerable benefits to our support team and, by extension, to our customers !
A precious help for our support team!
Behind every request is the NEO-FUGU® support team. Present on a daily basis, they deal with all customer requests, listen carefully to the problems encountered and endeavour to provide solutions as quickly as possible. Each ticket created corresponds to a request or difficulty raised by a user, and this work requires rigour, anticipation and responsiveness.
The team handles a large volume of tickets every day, and must constantly optimise the management of these requests to guarantee a smooth and efficient service. It’s a key role that contributes directly to customer satisfaction and the smooth running of our services.
The AI model developed responds directly to everyday problems:
🔮 Forecast the volume of upcoming tickets over several weeks, providing a clear vision of future needs, including recruitment requirements.
📌 Segment requests into specific categories such as development and parameterization, making it easier to manage priorities.
🔝 Managing and optimising our workload/capacity, to guarantee constant availability and responsiveness, even during periods of high activity.
This AI analysis enables us to anticipate peak workloads in order to optimise the management of the support team and guarantee service quality and resolution times that are as accurate as possible. As a result, our team is more responsive and even better organised to deal with requests! The progress and changes made will be shown on the dashboard that guides their days :
Pssst…No confidential information is divulged here! This is fictitious data, so don’t panic 😉
What technologies are we using?
At NEO-FUGU®, technology is not a joke !
Several tools were very useful to us during development :
- Python: the main programming language.
- Streamlit: for an interactive user interface.
- Prophet: a prediction model based on Machine Learning.
- Scikit-learn: for performance evaluation with Mean Absolute Error (MAE).
- Pandas: for data handling and analysis.
On vous explique comment ça fonctionne :
- Data loading: the user imports a CSV or Excel file containing ticket history.
- Handling and preparation: data is cleaned, formatted and organized by week and ticket type.
- Modeling with Prophet: the algorithm generates a forecast model for each ticket type.
- MAE calculation: comparison of predictions with historical values to measure accuracy and limit errors.
- Display of results: visualize forecasts and download predicted data.
And here's the result !
Thanks to AI, our support team can predict the future of customer requests, not in a crystal ball, but in the data! 🚀 The result? Less stress, more anticipation, and always a top-notch service. ✨