Staplin and Gish (2005)

1
Staplin and Gish (2005)

• Potential impact of driver turnover
• Replacement costs, etc.
• Greatest impact may be the area of safety (higher
  crash rate)
• Previous studies
• Some evidence that carriers with higher driver
  turnover has higher crash rates.
• Reliable estimate of the actual extent to which
  crash risk is increased as a function of
  job-change frequency is lacking.

2

• Goal of the paper
• Quantify risk function
• Test the statistical significance of the
  relationship between job change rate and crash
  rate
• Focus on crash data of for-hire interstate
  drivers.

3

• Method
• Analyze data from inspection records maintained
  by Federal Motor Carrier Safety Administration
  (FMCSA)
• Fatal, injury, and/or two-away crashes
• Job change rate is measured by the number of
  different carriers a driver has worked for in a
  given time interval.
• Use odds ratio to measure the relationship
  strength.
• Use two types of crash indicator (1) one or more
  crashes, and (2) two or more crashes.
• The latter is more likely to reflect drivers
  fault.

4

• Data
• Data errors removed
• Time interval should not be too short or too long
  (2 years or longer)
• Final data included 25,609 drivers, of which
  16,249 had no crashes, 8,797 had one, and 563 had
  two or more crashes.
• On average, drivers in sample had 6.1 inspections
  (ranges between 2 and 20).

5

• Results
• One job change more crash-free drivers
• Two or more job changes more crash-involved
  drivers.
• Odds of crash increases by 60 for drivers with 6
  or more job changes (no crash vs. one or more
  crashes).
• Odds of crash increases by 120 for drivers with
  6 or more job changes (no crash vs. multiple
  crashes).

6

• Results (cont.)
• Odds ratio increases sharply when a driver change
  jobs more than 2.5 times per year.
• Even more dramatic increase for drivers with 3 or
  more job changes per year.

7

• Discussion
• Proportion of crash-free drivers decline, without
  exception, as the job change rate increases.
• This relationship was most evident when the
  analyses focused on multiple crash data.
• If a driver changes jobs more than twice per
  year, the odds of crash begin to increase
• If changes job more than three times per year,
  crash odds is more than twice as high as those
  drivers with fewer job change rates.

8

• Questions
• Given this study result, what should you, as a
  carrier manager, should do.
• How do you balance the safety and driver turnover
  issues?
• What are the lessens from this study for
  shippers?
• What are the lessens from this study for policy
  makers?
• Does this study result give you some implication
  on how you play simulation game?

9
Mejza et al. (2003)

• Prior studies of truck safety
• Effect of carrier behavior on safety performance
• E.g. effect of noncompliance with HOS regulation
  on safety performance
• Present study
• Investigation of the characteristics of best
  safety performers.
• ID best performers
• Learn safety management used by best performers

10

• Focus
• Driver selectivity
• Instructional intensity
• Supportive and motivational intensity
• Sample
• Use SafeStat of FMCSA to list best 20 safety
  performers in each commodity category
• Surveyed 192 carriers
• 141 usable sample data
• Results

11

• Management Implications
• Be selective
• Training (multiple)
• Use multiple rewards

12

• Discussion questions
• If you are the carrier having poor safety record,
  what would you do?
• Do you rather hire drivers that have safe records
  or do you rather hire drivers that are likely to
  stay longer?
• Quality drivers may demand higher salary than
  others. Do you still want to hire them?
• If drivers are not driving safely, what impact
  does it have on carriers?
• What role does safety play for establishing
  relationships with shippers and consignees?
• Why is it important to survey the non-safe
  carriers too?
• What other factors would you consider as
  determinants of safety?