This year I’ve been learning Python and exploring its power to address complex challenges in contact centre planning. I’ve found that Python enables customisable solutions that allow you to go beyond the capabilities of Excel or Workforce Management (WFM) systems. Here are three areas where Python can make a strong impact:
Forecasting
Python enhances forecasting by enabling discovery and experimentation with a very broad range of statistical and machine learning models, supported by powerful data manipulation and processing tools. It simplifies the comparison of different candidate models.
Additionally, it helps automate the forecasting process, from data preparation to forecast output. It also allows for the production of charts programmatically, eliminating manual effort and ensuring consistency and efficiency.
I’ve built many statistical and machine learning forecasting models within Python now and have been delighted by the extra automation and accuracy that can be achieved.
Queueing Theory (eg ErlangC)
Python is ideal for automating contact centre resource requirement calculations from short-interval forecasts. I encoded the ErlangC formula into Python earlier this year and it performs exceptionally well. It calculates resource requirements for a large number of intervals with speed and reliability, taking into account process durations and target service levels (eg 80% of calls answered within 30 seconds).
Python offers the flexibility to explore other queueing theory models, or to customise the ErlangC formula to suit your specific operational needs – all achievable directly within the code. This adaptability makes Python a powerful tool for the accurate calculation of resource requirement and the optimisation of customer operation staffing.
Experimentation using simulation
With Python, you can model your customer operation as a computer simulation, where tasks like calls or emails are generated similarly to observations in real-life operations. These tasks are processed within the code, and the results can be validated against actual outcomes to ensure the model’s accuracy.
I developed a simulation for a contact centre using Python, which provided excellent results. This approach allows me to experiment with different routing strategies, skills assignments and staffing levels without making changes in the real-world environment.
While ErlangC and traditional queueing models are effective for simpler scenarios, simulation works well in handling routing complexities, delivering reliable insights for such cases.
======
That’s just three areas. There are many more, for example with scheduling and reporting. This is an exciting area for research and development.
Philip Stubbs is Partner of Atlantic Insight, and has over 25 years’ experience of improving performance within operational areas within a wide range of industries.
At Atlantic Insight, our mission is to help business operations achieve sustained improvements in operational effectiveness and customer engagement. We would be delighted to partner with you to improve performance. To start a conversation, please email us at hello@atlanticinsight.com, hit the Contact button on this website or call us on +44 (0) 161 438 2009.