KDnuggets Home » News » 2015 » May » Publications » Data Science for Workforce Optimization: Reducing Employee Attrition ( 15:n16 )

Data Science for Workforce Optimization: Reducing Employee Attrition


Predictive analytics is growing its reach, see how it is affecting workforce analytics domain. In this presentation Pasha Roberts explains what is in it for students, managers and practitioners.



By Mike Kennedy, Talent Analytics.

Predictive methods are often seen in marketing, finance, and logistics.  The use of these advanced methods for workforce analytics are an emerging and powerful domain for practice, with applications in talent selection, sourcing, training, development, planning and optimization.

This presentation by Talent Analytics Chief Scientist Pasha Roberts will provide an overview of the advanced analytics field with a range of applications that are emerging in application to a company’s largest and most important expense: their workers.

The presentation will review several case studies of talent analytics in practice.  Predictive concepts such as survival analytics, uplift modeling, and ensemble models will be applied to tangible customer situations.  In one case, a major financial services call center used predictive modeling to reduce attrition by over 30%, yielding a multi-million dollar savings.

  • Analytics students will hear “front-line” stories from real analytics engagements, where theory meets the constraints of real business practice.
  • Analytics managers will leave this session with a concrete understanding of the business value of talent optimization, showing how correlating “raw talent” to employee attrition can yield unprecedented results.
  • Analytics practitioners will leave this session with a concrete understanding of the analytics approach and models used to reduce attrition.

Here is the direct link to watch the presentation on Vimeo

Recorded at INFORMS / American Statistical Association Boston Chapter presentation at Bentley University,  February 24, 2015

Here is a report on this presentation:

Pasha Roberts, Talent AnalyticsOn the evening of February 24, a small but enthusiastic group from the Boston Chapters of INFORMS and the American Statistical Association gathered at Bentley University, along with Bentley University analytics students following online, to hear Talent Analytics Chief Scientist & Co-founder Pasha Roberts discuss the application of data science to workforce optimization.

After being introduced by Victor Lo, himself a pioneer of uplift modeling, Pasha explained the application of predictive concepts such as survival analytics, uplift modeling, and ensemble models to tangible customer situations. As Victor pointed out, this topic hadn’t been addressed at prior INFORMS Boston events, which prompted an excellent discussion with outstanding questions from an inquisitive audience of analytics and OR professionals.
Victor Lo, pioneer of uplift modeling.
In addition to providing many examples of predictive techniques, Pasha highlighted a detailed case study on using predictive modeling to reduce attrition by over 30%, yielding a multi-million dollar savings.

Overall, analytics managers and students walked away from the evening with a concrete understanding of the business value of talent optimization, showing how correlating “raw talent” to employee attrition can yield unprecedented results.

Related:


Sign Up