Session Outline
Customer Service Workforce Management Team forecasts are used in call-center’s agent scheduling, recruiting, and hiring of consultants on ad-hoc basis (approx. 260-300 agents). Forecasting is key in providing good service level to customers and balance operational workload in a cost-effective way. With the implementation of end-to-end ML model, customer service in Sweden went live in 2021 with predictive analytics in their daily operations
Key Takeaways:
- Worked very closely with business process experts to understand the current state and pain points, and desired target future state of the use case. Considered end-to-end implementation and maintenance process as well and training needs, change management need in way-of-working – these steps led us to true business value realization.
- Implemented a holistic view to understand data requirements of data engineering team and data science team, prepare scalable architecture for automated data onboarding, data processing, data exploration and then apply several ML models to identify the best model, used DevOps practice from day one (DevOps + MLOps), automated test scripts within MLOps – had an agile team comprises of business SME, Data engineers, DevOps Engineer, Data Scientists, Architect.
- Reduced the gap between data engineering and data scientists by onboarding them to the same data platform, emphasized on end-to-end automation.
Add comment