Predictive Analytics: Predict Driver Turnover

Handsome man truck driver standing by the truck

Just a couple years ago, who would have thought Predictive Analytics could forecast accidents? Acts of self-fulfilling prophecies being neutralized beforehand based on foreknowledge is simply too science fiction to be true, if not impossible all together. Victors of pragmatic hope, capacities of digital prophecies are bearing fruit in the most interesting ways. A few years ago it was used to accurately predict accidents of particular drivers based on flagged unsafe driving patterns. With the aid of advanced sensors today that are able collect even more types of data, the ability is fine tuned for likelihoods in even more specific ways.

As drivers push longer days on the road to expedite next day deliveries and A1 itineraries, so does pressure and stress that swells with the approaching dawn. Such negative health-efficiency correlation explains why freight providers struggle to streamline optimal business operation models with driver-friendly protocols.

Covenant Transportation Group, Inc. (CTG), a major US freight transportation and logistics services company headquartered in Chattanooga, Tennessee, that ships enormous volumes by land and water but endures significant rates of driver turnover despite extensive care programs in place, including Commercial Driver License Class A that is tailor designed for truck drivers. Such turnover factors in terminated contracts due to substandard performance, as well. CTG, thus, has put focus on enhancing Key Performance Indicators (KPI) as a counter measurement and key success factor to rising above industry standards. Another area of focus is the application of Predictive Analytic models to predict driver’s turnover.

Since inception in 2012, the company has realized practical benefits that derive from Predictive Analytics that can be directly applied to business operations. To enhance internal capacity in other departments, as well, the company sought subject matter expertise from Chris Orban with 10-year predictive computing experience under the belt. A key application is, nevertheless, is SAP Business Suite’s Enterprise Resource Planning (ERP). ERP is a business-management software that integrates a range of core applications required, collectively, to collect and interpret Big data, e.g., cash flow, raw materials and production, from various business sources for predictive purposes. Forecasts pertinent to different functionalities are thus made available and automated across departments and all business functions, accordingly. A centralized management system is used to run different functions based on a common database concept.

Different models apply different algorithms accordingly to distinct predefined segments of candidate pools, demographics and related criteria. For example, the model for drivers with under 6-month experience will be different from that of experienced drivers with 5 years of experience, one-day drivers differs from that of several days on the road, and etc. Several variables that are used to develop a model. Key variables can be duration and frequency of transporting missions, utilization rate, compensation in recent transport periods, as well as calculated stress levels likely to increase by drivers not knowing the schedule in advance.

As change in performance patterns are indicative of specific types of causes, CTG is also applying predictive capacities to identify likely roots of substandard behaviors, as well. For example, request for advance payment by drivers with no history of submit such applications is indicative of additional stress they might be undergoing, stress of which may increase the chances for accident or turnover, or not. The ERP model can algorize and translate hundreds and even thousands of variables in a blink of an eye for each forecasting model. A major advantage of the capacity is that it can evaluate viability of new variables and automate them in a streamlined and accurate manner.

In addition to financial factors, safety factors also affect drivers’ turnover rates. Upon being identified for security risks, drivers are obligated to undergo supplemental training programs or retain coaching capacities from certified safety personnel. Poor coaching capacities by management will likely endure higher turnover rates. With good resources readily available, on the other hand, the drivers’ willingness to open up to good coaching or participate in valuable training is a key success factor in reducing turnover. In the same way, refusal to embrace necessary help can be a determinant factor for drafting a notice of dismissal. CTG also discovered that drivers seeking employment else where while on company payroll, of which are included under the same criteria to receive a pink slip.

Enhanced Leadership:

When SAP predicts a driver is about to resign, a new standard operating procedure is for the manager to approach the driver to hold a casual dialog, such as “How is everything? Are you feeling ok?”. A recipe for disaster, managers that site overwhelming workloads are typically unlikely to hold casual conversations with team members under normal circumstances, i.e., as equals without asserting ‘I’m the boss’ tones or siting the alphabet soups in front of the name when out of context. The procedure has proved to be a doubled benefit and now CTG has initiated a new requirement for fleet managers to spend 5-10 minutes engaging with the team members to boost morale and rally the troops as well as learn about the lives of people around them. Communication is an excellence that prerequisites teamwork. Without it, managers will be walking the tightrope of team-wide inflexibility as a result of treating members as robots and not knowing what is going on nor how to compensate for flexibility. The difference starts with the view of team members as either capable subject matter experts or simply subordinates. The latter, expecting professionalism is a losing battle before entering the ring. The former, not knowing how to handle ownership that comes with competency being a better scenario of the two but, however, still a rough ride to deliver. Rule of thumb among world class organizations: team wide dysfunctionality points to the manager in charge, first. Leading by example takes objectivity, pride and passion; thus, professionalism, yet, humility and gratitude for the help provided by the team. With trust, let-goes are justifiable with, yet, maintaining a long term relationship. The workforce has a right to determine appropriate levels of effort according to provided wages and should there be no common ground, it is their right also to leave. Leadership selection, nevertheless, is one of the most fundamental and complex process among world class organizations, if not a corporate culture impossible to duplicate, altogether. Sometimes, a touch of compassion can amount to unimaginable operational efficiency.

Predictive models, nonetheless, can be used to identify management technigue that work, as well as the ones that don’t, and aid decision making capacities. Predictive Analytics are expected to reduce driver turnover rate by 5 to 8 percent. That is monumental for keeping the good ones.

*** Please comment (Log-in required), Like, or share the article on facebook below. ***

Compiled by BLOG.SCGLogistics

Reference and photos

Share this post