Advanced predictive analytics for employee turnover

点击翻译按钮获取中文摘要

In the current era of big data, high volumes of a wide variety of valuable data about employees can be easilycollected at a high rate in various industries (e.g., transportation & trucking industry, airline industry, certaintypes of manufacturers & service companies with highly trained/accredited employees). Data sciencesolutions--which apply multidisciplinary techniques such as data mining, machine learning (including deeplearning), and statistical modelling--help discover implicit, previously unknown and potentially usefulinformation and knowledge that is embedded in the big data. Our industrial partner, DecisionWorks ConsultingInc, has reported that the issue of driver turnover is significant for its clients in the transportation & truckingindustry because it costs roughly $5K to train a new driver. Furthermore, bad drivers (e.g., those who are hardon equipment, have poor safety records, or who find driving under anything but good road conditions difficult)are causing problems such as late or mismanaged deliveries, higher than average maintenance costs, etc.Similar issues exist in other industries (e.g., airline industry, manufacturers & service companies with highlytrained/accredited employees) where the cost of turnover on a per-employee basis is high. Having data sciencesolutions to discover knowledge and information about employees therefore helps reveal the characteristics of"good" and "bad" employees, which in turn helps DecisionWorks to enhance its business analytic model for (a)reducing operating costs and (b) enhancing the employees' working environment for its clients. With success inour Engage project in predictive analytics of turnover of a specific type of drivers in transportation & truckingindustry, we plan to further enhance our data science solution for predictive analytics of (a) other types ofdrivers and (b) employees in other industries. Moreover, we also plan to design and develop a visual analyticscomponent for our data science solution so that the discovered knowledge can be represented in a visual formthat can be easily comprehend. The resulting solution is expected to perform effective visual analytics andaccurate predictive analytics to forecast good employees at risk of turnover.