In Uncategorized

Common driver frustrations centered on safety, health and nutrition, pay, and long hours with little sleep can add up and lead a driver to quit or perform at less than optimal levels. There is no one formula for success in driver retention, but surprising answers lie in the masses of data that your fleet collects.

Common driver frustrations centered on safety, health and nutrition, pay, and long hours with little sleep can add up and lead a driver to quit or perform at less than optimal levels. There is no one formula for success in driver retention, but surprising answers lie in the masses of data that your fleet collects. Data can help fleet managers understand specific frustrations of their workers so they can prevent resignations or driver burnout.

These are the reasons why fleets need a unified view of critical information on driver performance, the ability to analyze driver issues, and to input remediation conversations that allow for searchable and documented driver communication. What if you could tell when and why a driver is frustrated? Data and analytics are at the heart of the solution that can help your fleet drastically improve safety and productivity, as well as keep drivers happy with effective remediation and full visualization of driver issues.

Identifying at-risk drivers with structured and unstructured data

Fleet managers can take advantage of data to learn about driver frustrations and indicators that a driver is going to resign. There are two main types of data that managers can look at — structured and unstructured.

Structured data is used more than unstructured data for this sort of analysis today. It could be data from databases, XML or web service files, a mainframe system such as AS/400, spreadsheets, Access database, or any cloud-based system. Examples of structured data are pay rates, miles driven, on-duty hours, and service failures. Managers often access structured data through a website or employee directory.

Unstructured data includes images, such as an image from an accident site. Images convey more information than what one can read or be told by someone. Social media messages are another example of unstructured data. Some fleet managers monitor social media groups to address any issues raised by their drivers.

Video files are another form of unstructured data. In trucking, forward-facing dash cam videos can tell fleet managers a lot of information such as whether the driver was following too closely or if the car in front of the vehicle was at fault in an accident.

In-cab messages between drivers and managers sent through telematics devices are another example of unstructured data. Like a pager for trucks, in-cab messages are conversations between truck drivers in the field and fleet operators or driver managers in the back office. Data that can be pulled from these messages include morning check calls before driving starts, messages about driving trucks empty, shipper and home arrival messages, and the numbers shown when fueling trucks. In-cab messages can be customized by each carrier and therefore differ from one carrier to the next.

There is valuable data embedded within these images, social media messages, audio and video files, and in-cab messages. Today, most metrics and reports are based on structured data, but there is more and more value in the unstructured data.

Uncovering reasons for driver resignation through data

Two case studies based on anonymized data from real Omnitracs Analytics clients illustrate the value of data in driver retention and how the value increases as structured data is paired with unstructured data.

In each case study, the Omnitracs Analytics team mined data for both carrier customers to understand why each driver quit. To do this, the team removed frequent words (‘the,’ ‘from,’ ‘for,’ etc.) from in-cab messages and then plotted the most commonly-occurring words (‘gallon,’ ‘fuel advance,’ ‘home,’ etc.). Plotting these in-cab messages creates a breadcrumb trail of the driver’s work day, showing when check calls, arrivals and departures, time in the shop, and empty driving time occur. The team then gets a high-level perspective using text mining tools. From there, reviewing the actual messages paints a picture of the drivers’ frustrations.

Case study 1: Service breakdowns lead to empty miles — and resignation

In this first case study, a good driver spent three years with a trucking company before leaving.

Looking at the driver’s messages, the Omnitracs Analytics team found frequent service breakdowns. For example, the driver’s trailer would be down one week and the next week there would be an issue with the truck.

Examining his in-cab messages, the team was able to see that this driver was sending large amounts of messages with the word ‘home.’ These messages were either messages written to or from the driver or they were geo-fencing-based messages related to when the driver arrived at or departed from his home. The word ‘home’ appeared 88 percent more than average with this driver and he had empty and multiple service failures in a 30-day period.

While this driver’s pay was eight percent above other drivers, his miles were down by 106 percent. A bonus or guaranteed pay program could be an explanation. The driver’s on-duty and driving hours, calculated based on hours of service logs, were also down. Reviewing the logs, one can monitor when shifts change from nine to five to five to nine in a week. This particular driver had an 8.8-hour difference in his shift timing — a moving average of his weekly miles. This moved down several times, presumably due to service failures when his truck was in the shop.

Looking at unstructured data tells you the whole story. This driver was not driving enough miles and he was driving empty. Even though he was still making pay, he was probably unhappy that he was not driving enough miles or making enough money.

Case study 2 – Driver pay dips and increased visits home lead to resignation

The second case study is about a driver who was with the carrier company for ten years and was a great employee but left without warning.

The carrier company’s average number of messages from driver to manager was 63.1 and this particular driver sent an average of 353. The word ‘home’ was present 87 percent more than average in this driver’s in-cab messages.

Plotting all of the driver’s trip locations and destinations on a map, one can analyze his miles and where he has traveled. Looking at structured data, his weekly pay was down 23 percent, his miles were down 13 percent, his on-duty hours were down 54 percent, and his on-duty driving was down 21 percent. There was also a huge drop in the driver’s pay the week before he resigned.

The reason, it turned out, why this driver left the company was because his wife was ill. The driver wanted to be home more often to take care of her. If the carrier company could have seen this ahead of time, the manager could have potentially offered an unpaid leave or given the driver time off. Instead, the company lost a good, tenured driver.

A single, intelligent system for fleet data

When fleets combine all data from human resources, operations, accident data, trip pay, and safety data into a single system, it’s easier to look for trends.

And when they combine unstructured and structured driver data, they can glean better insights into how aspects such as personal issues or service failures can affect driver outcomes.

Omnitracs Driving Center, part of the Omnitracs Analytics platform, is a useful tool for this that pulls together different data elements to provide a unified view of driver performance along with the ability to spot trends, analyze issues, and document improvement efforts.

The Driving Center simplifies the process of finding the right data in a usable format. It gives fleets a unified vision of driver information related to safety and compliance, and the ability to identify the most at-risk drivers and measure the impact of safety and coaching programs.