Every process managed by humans is bound to have at least a small error rate. After all, as Alexander Pope wrote, “To err is human.” Technology has played an important role in developing processes resilient to human error, yet human error persists.
In some cases, it seems that technology has overcompensated for human errors and resulted in unintended consequences. Research from an ASCM-published journal found that monitoring technology intended to prevent truck drivers from exceeding their maximum daily hours of service instead caused drivers to speed and thus drive unsafely in order to drive farther during their allowable driving time.
The authors of this article wanted to study this phenomenon further in order to determine the impact of frontline employee errors on deliveries. Like all people, drivers and dispatchers make mistakes. Because supply chains are so tightly coupled, even a minor error can have adverse consequences. Of course, these errors are not necessarily due to lack of effort or concentration. Instead, they are often side effects of the interaction between humans and technology.
To prevent driver errors, carriers invest in sophisticated technologies such as speed regulators, collision mitigation systems, biometric fatigue sensors, video-monitoring systems, rollover stability systems, geo-fencing systems and lane-departure warning systems. Furthermore, since late 2017, the Federal Motor Carrier Safety Administration has required onboard electronic logging devices (ELDs) that report drivers' speed, idle time, hard-braking behavior, vehicle location, engine operating hours and vehicle miles at frequent intervals. Many carriers have embraced this technology, voluntarily exceeding the mandate to ensure the safety of their drivers, vehicles, cargo and the public.
To prevent dispatcher errors, carriers and shippers invest in transportation management systems (TMSs) to help plan shipments and routes, determine carrier mixes, match cargo with vehicles and track shipments. Although TMS technology has provided powerful managerial tools for decades, it is not infallible.
When leveraging these technologies, unintended consequences can arise, including over-trusting a technology, overriding information not perceived as beneficial and not being as vigilant because it is believed the technology will flag issues. This makes humans the weakest link in these supply chains.
As one shipping company manager noted during study interviews: “Having a $10,000 computer that's hooked up to a $60 million GPS system doesn’t eliminate all problems. There is still a person behind the keyboard who is acting on the information or relaying that information. I wish I could control or predict that person, but human nature just kicks in.”
Furthermore, the likelihood of these errors could depend on latent conditions — or systemic managerial, technology and social conditions that define a firm. These can include hiring and training practices, policies, priorities, equipment, information technology systems, norms, values and safety culture. Latent conditions that intensify the negative effects of a human error are known as resident pathogens, while latent conditions that reduce the negative consequences of a human error are known as defensive layers.
For example, a trucking policy such as no pay for out-of route miles could incentivize making illegal U-turns to compensate for navigational errors. Similarly, mileage-based pay could incentivize drivers to work in hazardous weather conditions. A defensive layer could be to financially reward drivers for safe driving practices.
There also are technology resident pathogens. For example, if a TMS does not have a sufficient amount of built-in checks that watch for dispatcher data entry errors, a driver might end up having to speed, make an illegal turn or make a quick lane-change to compensate for the errors. Similarly, the inability of a carrier to compile and analyze data from the onboard ELD or track-and-trace system prevents dispatchers from realizing when unsafe driving occurs.
Details in the data
So, then, how do latent conditions at transportation companies intensify or reduce the likelihood of human errors and late deliveries? To answer this question, consider shipment-level data that was collected from the TMSs of a Fortune 500 company (LIM). This business produces furnishings, appliances, consumer electronics and housewares and has an annual revenue of more than $20 billion. LIM's freight and warehousing expenses comprise 16% of the company’s revenue. Prior to data collection, LIM had invested more than $1 billion in advanced systems and technology from leading vendors and installed next-generation GPS tracking technology that updates every 30 seconds on every truck. However, despite substantial investments in technology defensive layers, some shipments continued to be lost, delayed or routed to incorrect locations. In fact, more than 5% of LIM’s deliveries — about 14 per day — were delayed because of errors by dispatchers or drivers. Thus, every few hours, LIM needed to address a disruption in its 24/7 operations. This is a shockingly large number of errors, particularly when a company already has made substantial investments in error-prevention technology.
Nearly 300,000 53-foot-trailer shipments, conveyed by 97 carriers from 2014 to 2016, were examined. Although looking at just one company’s data limits the generalizability of any findings, it also removes more variables, which means the findings also are stronger. It also is important to note that there could potentially be human errors in the entered data that could have affected the results, but it’s impossible to know the rate of error because that is not reported.
Included in the study was data about the artifacts or signs of carrier latent conditions — which includes unsafe driving, hours-of-service (HOS) violations, vehicle maintenance and driver fitness violations. This information was collected about LIM’s carriers from the Federal Motor Carrier Safety Administration’s Compliance, Safety and Accountability (CSA) program dataset. A carrier’s CSA score is the sum of its violations, as ticketed or reported from roadside inspections, weighted by the severity and recency of the violations. Scores range from zero to 30, with zero being the best.
Leveraging this data, driver and dispatcher errors, as well as the carrier latent conditions described above, were analyzed. The dependent variable was the likelihood that a shipment would be delivered late (at least 15 minutes after its scheduled arrival time). The hypothesis was that driver and dispatcher errors each increased the likelihood of a late delivery, and the likelihood of a late delivery would be enhanced if the shipment was managed by a carrier with more CSA violations.
Naturally, the data showed that the presence of a dispatcher error increased the likelihood of a late delivery. In fact, deliveries associated with a dispatcher error were significantly more likely to be late than those not associated with a dispatcher error. The same is true for driver errors. In addition, the average length of delay for late deliveries was longer for dispatcher errors.
When reviewing the interactions of latent conditions with these errors, carriers with a record of more HOS violations were more likely to have dispatcher errors and late deliveries. There was some influence from unsafe driving and vehicle maintenance violations, but this influence was not significant. Driver fitness violations also had a slight impact.
For driver errors, the cross-level interaction between a driver error and the carrier’s record of HOS violations was positive and significant. There also was significant cross-level interaction between a driver error and driver fitness violations.
The length of delay for dispatcher errors, compared with driver errors, was longer for shipments transported by carriers with a record of more HOS and vehicle maintenance violations, respectively.
Another interesting finding was that only driver fitness violations were linked to a reduced likelihood of delivery delays. This means that there is no benefit to unsafe driving, HOS violations or vehicle maintenance violations because they do not improve the chances of a delivery being made on time. It’s better to be safe than cut corners.
A route to improvement
Focusing directly on preventing errors, such as through the use of technology that can reduce human errors, is not enough. Instead, organizations must examine the latent conditions that contribute to higher error rates. In many cases, they can only be addressed by managers, rather than individual workers.
Management can implement new defensive layers or strategies to reduce human errors that lead to late deliveries. For example, by instituting platooning driving practices, a carrier can have four drivers work together to double-check directions and keep an eye on the other drivers’ behavior. Similarly, two massive mega-trucks carrying the same amount of cargo as three standard trucks have proportionally fewer opportunities for human errors. Other managerial defensive layers include working only with carriers that have well-integrated technology-enabled systems that enable 360-degree visibility and have a good driving record.
Investment in technology still is a good idea, even if it does not eliminate human error. New technologies improve visibility and provide updated arrival time information, allowing a dispatcher to adjust recipients' expectations. Other technologies allow validation of vehicle type, licenses and permits prior to dispatching. Artificial intelligence (AI) and autonomous vehicle technologies reduce the adverse consequences of errors because automation replaces drivers and dispatchers. Predictive analytics, including AI systems, can reduce the impact of dispatcher errors.
Until more defensive layers are implemented, resident pathogens will cause more late deliveries in the future. It is bad enough that so many violations occur naturally and because of resident pathogens, but what makes it even worse is that cutting safety corners appears to compound the adverse consequences of errors, rather than leading to faster deliveries. This emphasizes the importance of investments in technology and reassessment of policies and priorities by carriers. In the meantime, shift the focus to safety.
ASCM members can access research from an ASCM-published journal.
Take a deeper dive into this topic at the Distribution that Goes Beyond State-of-the-Art educational session at the 2022 ASCM CONNECT Annual Conference on Sunday, September 18.
