Putting data to work: How to optimize workforce staffing to improve organization profitability
New machine learning technologies allow companies to apply better staffing strategies by taking advantage of historical data. Francesca Lazzeri and Hong Lu share a workforce placement recommendation solution that recommends staff with the best professional profile for new projects.
Talk Title | Putting data to work: How to optimize workforce staffing to improve organization profitability |
Speakers | Francesca Lazzeri (Microsoft), Hong Lu (Microsoft) |
Conference | Strata Data Conference |
Conf Tag | Make Data Work |
Location | New York, New York |
Date | September 26-28, 2017 |
URL | Talk Page |
Slides | Talk Slides |
Video | |
The recent advancement in machine learning and big data technologies allows companies to apply better staffing strategies by taking advantage of historical data. Assigning the right people to the right projects is critical for both the success of each project and the overall profitability of the organization. Most commonly, project staffing is done manually by project managers and is based on staff availability and prior knowledge of an individual’s past performance. This process is time consuming, and the results are often suboptimal. This process can be done much more effectively by taking advantage of historical data and advanced machine learning techniques. Francesca Lazzeri and Hong Lu share a workforce placement recommendation solution developed for professional services company Baker Tilly Virchow Krause that recommends staff with the best professional profile for new projects. By aligning staff experience with project needs, the solution helps project managers at Baker Tilly perform better and faster staff allocation, with the final goal of improving Baker Tilly’s profit. Based on an offline evaluation, a 4~5% improvement on profit is expected for the projects employing the solution. The solution has been integrated with Baker Tilly’s internal practice management system and will be evaluated in a few pilot teams before being implemented across all teams. Francesca and Hong offer an overview of the solution and experiment design, which predicts staff composition and computes a “Staff Fitness Score” (rating) for a new project; explore the solution architecture—new projects in Baker Tilly’s database are processed daily by the Azure Machine Learning web service, the results are consumed by project managers in Baker Tilly’s practice management system, and the workforce placement recommendation results are also visualized on a real-time PowerBI dashboard; and explain how they implemented an offline evaluation of the solution using machine learning models (gradient boosting trees) to predict the contribution margin per hour (CMH) of 5,000 randomly sampled projects using the recommended staff and comparing the predicted CMH with actual project CMH to estimate the potential improvement in CMH.