Impact of Behavior-Based Safety Techniques on Commercial Motor Vehicle Drivers (2007)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Method The main tool for collecting information on fleet safety managers’ opinions on BBS techniques was a survey. The sur- vey was distributed to safety managers through several dif- ferent methods, including (1) a secure Internet survey form, (2) a survey form completed on the computer and returned via email, and (3) a traditional paper-and-pencil survey form returned via facsimile or mail. Appendix B shows the computer and paper-and-pencil survey forms. This section describes the survey methodology in more detail, and the next section of this chapter provides principal results. Survey Design and Content The project team conducted two focus groups to identify specific behavioral safety-management techniques currently used by CMV carriers. One focus group was held at the annual Virginia Trucking Association Safety Managers meeting in Williamsburg, VA, while the other focus group was held in Knoxville, TN, at a local CMV operation. The discussions addressed targeted behaviors and performance measures (including those captured by OBSM devices) and behavioral safety-management practices used by carrier safety man- agers. These practices included prompts (e.g., signs, posters, reminders) and consequences (i.e., rewards and punishments) among other behavioral interventions. As BBS encompasses a variety of different techniques, procedures, and terminology, these focus groups were critical in identifying current behav- ioral management practices in CMV operations (including the terminology safety managers use in describing these prac- tices). Analogous to the results found in the LTCCS, most of the at-risk driving behaviors identified by focus group respondents were decision and recognition errors (see “Prin- cipal Survey Results, Part 1: Safety-Critical Behaviors”). The focus groups confirmed the project team’s hypothesis—that few focus group respondents implemented a comprehensive BBS program. Further, focus group respondents indicated they use specific behavioral management techniques but did not identify these techniques as “BBS.” The information gleaned from these focus groups was essential in developing the survey for fleet safety managers, including terminology, specific behavioral management techniques, driving and non-driving behaviors, observation techniques, and barriers/problems in implementing BBS techniques. Commercial truck operations have very high driver turn- over rates, thus accentuating the need to implement effective safety-management techniques. Both the focus groups and survey questionnaire addressed how BBS techniques can affect retention positively or negatively. A practical reality is that most motor carriers will not implement techniques that have adverse impacts on retention, even if there are safety benefits. Appendix E displays an example of the focus group presen- tation. The focus group presentation was used as a guide for the presenter, but also as a way to elicit responses from the focus group attendees. The focus group discussions were conducted similarly to the four steps in implementing BBS techniques (described previously in “Behavior-Based Safety Principles”): 1. Identify safety-critical driving and non-driving behaviors 2. Perform observations to gather data 3. Provide feedback to encourage improvement 4. Use gathered data to identify factors promoting positive change The dialogue began with a discussion identifying the most critical safety behaviors (mostly driving behaviors but some non-driving behaviors such as loading/unloading), pro- ceeded to a discussion of how safety managers gathered infor- mation on these behaviors (e.g., ride-alongs, surveillance, OBSM devices), talked about ways to provide feedback and related contingencies, and then discussed management poli- cies and practices put into place based on this process. This Survey Method and Results 24

25 structure was simple and lent itself to a non-technical discus- sion of specific management practices. The focus groups included safety managers who have tried various specific BBS techniques and who were willing to pro- vide case-study information on their experiences for the proj- ect report. The emphasis was on successful applications but lessons may be learned from unsuccessful ones as well. Based on the literature and focus group discussion, the fleet safety manager survey form contained 48 questions. These were divided into five parts: 1. Safety-Critical Behaviors. Questions 1–22 listed safety- critical driving and non-driving behaviors. Safety-critical behaviors were those behaviors that impacted overall safety (e.g., injuries, incidents, and crashes). Using a 5-point scale (0 = No Relationship and 4 = Strong Relationship), respon- dents were asked to rate how strong the relationship was between each of the safety-critical behaviors with driver crash risk and non-driving-related illnesses and injuries. 2. Observation of Safety-Critical Behaviors. Questions 23–26 asked respondents to indicate if their organization currently used the observation technique listed. 3. Specific BBS Techniques. Questions 27–35 asked respon- dents to indicate if their organization currently used the specific BBS technique. Then, the instructions stated, “If ‘Yes,’ please rate its effectiveness.” 4. Barriers/Problems to Implementing BBS. This section consisted of seven questions (Questions 36–42). Using a 5-point scale (0 = No Barrier/Problem and 4 = Serious Barrier/Problem), respondents were asked to rate how strong the barrier/problem was in implementing BBS. 5. Comments/Respondent Information. Questions 43–48 asked respondents to comment on BBS and/or any ques- tions in the survey. Blank spaces were provided to write these comments; this information on the completed forms was reviewed separately and reported selectively in this synthe- sis. There were also four questions regarding survey respon- dent demographics: two questions on the safety manager’s years of personal experience and two questions on the size and operation type of his/her fleet. The last question in the survey asked respondents if the research staff could contact them directly to discuss their organization’s BBS programs. Survey Distribution and Analysis The survey forms were primarily distributed via email. Potential fleet safety manager respondents were identified from attendees at the 2005 International Truck and Bus Safety and Security Symposium in Alexandria, VA, and the 2005 Virginia Truck Association Safety Manager meeting in Williamsburg, VA. Other survey recipients were respondents from previous CTBSSP studies for whom contact information was available. Further, members of the National Private Truck Council were contacted. All survey forms were accompanied by a cover email explaining the survey. Altogether, approxi- mately 400 fleet safety manager survey forms were distributed. While the pool of potential survey respondents was gener- ated from a sample of North American respondents, the cur- rent sample is best described as a sample of convenience as survey respondents were self-selected. Project resources did not allow the research staff to obtain a representative sample of CMV operations. Note that all survey responses were con- fidential and no attribution is made to these responses unless permission was granted by those parties. Principal Survey Results Part 1: Safety-Critical Behaviors Respondents were asked to rate 22 safety-critical driving and non-driving behaviors with regard to their strength of association with crash, injury, and illness risk. The scale went from “0” (no relationship) to “4” (strong relationship); thus, the higher the number the greater the association between the non-driving and driving behaviors and risk (i.e., crashes, injuries, and illnesses). Respondents rated 12 non-driving safety-critical behaviors (such as poor lifting techniques, improper cargo securement, etc.) and 10 safety-critical driv- ing behaviors (such as careless backing, speeding, etc.). The highest-rated associations for driving behaviors were improper following distances (mean rating of 3.7), speeding (mean rating of 3.6), and high speeds on curves and ramps (mean rating of 3.6). As can be seen in Table 1, the highest- rated driving behaviors reiterate the results found in the LTCCS (i.e., decision errors were the most frequent critical reason). The lowest-rated associations for driving behaviors were inappropriate left turns and inappropriate right turns (mean ratings of 2.8 and 2.9, respectively). The highest-rated associations for non-driving behaviors were drugs and alcohol (mean rating of 3.3) and failure to inspect vehicle pre-/post- trip (mean rating of 3.2). Note also that these highest-rated items are non-driving behaviors but have relevance to driving safety. The lowest-rated associations for non-driving behav- iors were for improper docking of the truck, smoking, and poor diet (mean ratings of 2.1, 2.3, and 2.3, respectively). The safety-critical driving and non-driving behaviors, mean ratings (to the nearest tenth), and rankings are presented in order in Table 1. When there were ties in the mean ratings, rankings were determined by looking at additional decimal places. However, for simplicity, these are not shown in the table. As can be seen in Table 1, respondents rated the association between driving behaviors and crash and injury risk (ranging from 2.8 to 3.7) much greater than non-driving behaviors and injury and illness risk (ranging from 2.1 to 3.3). Most driving behaviors were rated as “moderately strong” to “strong,” while non-driving behaviors were rated as “moderate” to

“moderately strong.” These ratings make intuitive sense as fleet safety managers are primarily responsible for addressing the on-road safety of their employees. Support for this con- tention is found below. Part 2: Observation of Safety-Critical Behaviors Part 2 presented three questions regarding observation techniques used by respondents to monitor safety-critical driving and non-driving behaviors. If respondents indicated using the observation technique, they were asked several follow-up questions, which varied depending on the tech- nique. These follow-up questions concerned (1) the device the organization uses, (2) what behaviors the device tracks, (3) the frequency with which the organization conducts ride-alongs or covert observations, and (4) who conducts the ride-alongs. OBSM Devices Question 23 asked respondents if they currently use an OBSM device. Twenty-seven respondents (41.5%) indicated they currently use some form of an OBSM device to observe safety-critical non-driving and/or driving behaviors. Table 2 26 Fleet Safety Managers (n = 65) Safety-Critical Behaviors Mean Rank Driving Behaviors Improper following distances (i.e., tailgating) 3.7 1 Speeding (i.e., maximum cruising speeds) 3.6 2 High speeds on curves and/or ramps 3.6 3 Driving while fatigued 3.5 4 Careless lane changes 3.5 5 Disregard of traffic signals (e.g., stop sign, red light, etc.) 3.5 6 Careless backing 3.4 7 Attention to roadway (e.g., engaging in distracting activities) 3.4 8 Inappropriate right turns 2.9 9 Inappropriate left turns 2.8 10 Non-Driving Behaviors Drugs and alcohol 3.3 1 Failure to inspect vehicle pre-/post-trip 3.2 2 Behaviors that lead to slips, trips, and falls 3.1 3 Improper attachment of the trailer to the tractor 3.0 4 Poor lifting techniques 2.9 5 Improper cargo securement 2.7 6 Improperly entering/exiting truck 2.6 7 Failure to plan trip 2.6 8 Poor exercise habits 2.5 9 Poor diet 2.3 10 Smoking 2.3 11 Improper docking of the truck 2.1 12 Table 1. Mean degree of association with risk for safety-critical driving and non-driving behaviors. OBSM Device % Use (out of 27) Safety-Critical Behaviors Tripmaster® 4.4% Speed, hard braking XATA 21.7% Speed, brake applications Qualcomm® 43.5% Speed, hard braking Eaton VORAD® 8.6% Following distance, lane changes PeopleNet 4.4% Speed, hard braking DriveCam® 4.4% Speed, hard braking, inattention-related behaviors, fatigue Cadec Mobius TTS® 8.6% Speed, hard braking, erratic driving International Road Dynamics 4.4% Speed, hard braking Note: The devices used are based on reported survey results. The CTBSSP and TRB do not endorse products. Table 2. OBSM devices used by survey respondents and the safety-critical behaviors observed.

27 shows the different OBSM devices used by respondents and what safety-critical behaviors these devices observe and record. As can be seen in Table 2, most OBSM devices are used to track the speed and braking behaviors of drivers. Ride-Alongs Question 24 asked respondents if they currently conduct ride-alongs with their drivers. Thirty-eight respondents (58.5%) indicated they currently perform ride-alongs to observe safety-critical driving and/or non-driving behaviors. Figure 9 shows the frequency that respondents indicated per- forming ride-alongs. As can be seen in Figure 9, most respon- dents indicated performing annual ride-alongs with their drivers (28%), while few reported performing daily ride-alongs (3%). Table 3 displays what personnel conducted these ride- alongs. As can be seen in Table 3, most respondents indicated the driver’s supervisor (65.8%) is responsible for performing the ride-alongs. Most BBS programs suggest a co-worker be the individual performing the observation. Studies suggest peer observation increases camaraderie among workers, trust in the observation (as co-workers are more aware of the daily job difficulties than management), and fact finding rather than fault finding. The percentages in Table 3 do not sum to 100% as respondents were able to select multiple personnel. Covert Observations Question 25 asked respondents if they currently use covert observations (e.g., hidden camera or observers). Twenty-four respondents (36.9%) indicated they currently use covert observation techniques to observe safety-critical driving and/or non-driving behaviors. Figure 10 shows the distribution of how often respondents indicated performing covert observa- tions. As can be seen in Figure 10, most respondents indicated performing quarterly, weekly, and daily covert observations with their drivers (21%, 21%, and 20%, respectively), while few reported performing bi-annual covert observations (4%). Comments from the Public Question 26 asked respondents if they currently receive comments from the public (e.g., from 1–800 “How’s my Driving” phone service or from clients). Thirty-eight respon- dents (58.5%) indicated they receive and use comments from the public to observe safety-critical driving and/or non-driving behaviors. Overall, 54 respondents (83.1%) reported using some type of observation technique to observe the safety- critical driving and non-driving behaviors of their drivers. Note that 11 respondents (16.9%) reported using none of the observation techniques listed. If and how drivers in these organizations get feedback on their driving and/or non- driving behaviors is not known. Part 3: Specific Behavioral Management Techniques Part 3 presented nine questions regarding specific behav- ioral management techniques. Questions 27 to 33 asked fleet safety managers to first indicate whether they use the specific behavioral management techniques, and then, if “yes,” to rate its effectiveness in reducing unwanted or at-risk driving and non-driving behaviors. Further, if respondents indicated using the specific behavioral management technique, they were asked several follow-up questions, which varied depending Driver from Same Terminal Terminal Manager Driver’s Supervisor Independent Observer Other Manager % (of 38 “yes” respondents) 26.3% 36.8% 65.8% 15.8% 34.2% Table 3. Personnel who conduct ride-alongs. Daily 3% Weekly 11% Monthly 23% Quarterly 24% Bi-Annualy 11% Annually 28% Figure 9. Frequency of ride-alongs among “yes” respondents.

on the specific behavioral management technique. These follow-up questions concerned (1) the frequency with which feedback is given to drivers, (2) how feedback is given to driv- ers, (3) the type of feedback given to drivers, (4) the frequency with which the organization holds training and education sessions, (5) the types of rewards or penalties given to drivers, and (6) what data the reward or penalty are based on. Ques- tions 34 and 35 asked respondents to indicate the five driving or non-driving behaviors their organization focuses on dur- ing training of new drivers and refresher training with expe- rienced drivers, respectively. The specific behavioral management techniques, mean ratings, and rankings are presented in order in Table 4. When there were ties in the mean ratings, rankings were deter- mined by looking at additional decimal places. However, for simplicity, these are not shown in the table. Table 4 shows the percentage of respondents who use specific behavioral man- agement techniques and the effectiveness of the techniques. The scale went from “0” (highly ineffective) to “4” (highly effective); thus, the higher the number, the greater the effec- tiveness of the specific BBS techniques in reducing unwanted or at-risk driving or non-driving behaviors. The highest- rated behavioral management technique was training and education programs directed at specific driving behaviors (mean rating of 2.9). The lowest-rated behavioral manage- ment technique was driver self-management/self-observation (mean rating of 2.3). Most of the specific behavioral manage- ment techniques were rated as neutral to effective. Peer Observation and Feedback Question 27 asked respondents three follow-up questions if they indicated using peer observation and feedback. One follow-up question asked respondents to indicate the frequency of the feedback given to drivers when using this behavioral management technique. Of the 41 respondents who indicated using peer observation and feedback, 14.6% gave daily feed- back, 12.2% weekly, 34.2% monthly, 22% quarterly, 2.4% bi- annually, and 14.6% annually. There is consensus in the literature indicating the frequency of feedback is related to the effectiveness of the feedback. That is, feedback given more often will produce more behavior change than feedback given less often (Alvero, Bucklin, and Austin, 2001). Table 5 shows the effectiveness of peer observation and feedback by the fre- quency of feedback as indicated by survey respondents. As can be seen in Table 5, survey responses did not vary enough to support or refute this hypothesis. Another follow-up question asked respondents to indicate how often feedback was given to drivers. Of the 41 respondents who indicated using peer observation and feedback, 96% of respondents reported giving drivers feedback via a one- on-one meeting with the safety manager, 48.8% via private 28 Daily 20% Quarterly 21% Bi-Annually 4% Annually 17% Monthly 17% Weekly 21% Figure 10. Frequency of covert observations among “yes” respondents. Fleet Safety Managers Specific Behavioral Management Techniques % Who Use (out of 65) Mean Rank Training and education on specific driving behaviors (e.g., mirror use, lane changes, following distance, etc.) 100% 2.9 1 Peer observation and feedback 63.1% 2.8 2 Disincentives/punishment 87.7% 2.6 3 Training and education on specific non-driving behaviors (e.g., lifting techniques, diet, exercise, etc.) 72.3% 2.6 4 Incentives/rewards 80% 2.5 5 Prompts 89.2% 2.4 6 Driver self-management/self-observation 32.3% 2.3 7 Table 4. Use of specific behavioral management techniques by survey respondents and the effectiveness and rankings of the techniques.

29 memo/letter, and 24.4% via public display (note the sum does not equal 100% as respondents could select more than one choice). The last part of Question 27 asked respondents to indicate the type of feedback given to drivers. Of the 41 respondents who indicated using peer observation and feedback, 39% of respondents reported giving drivers indi- vidual feedback regarding their driving and non-driving behaviors, 7.3% reported giving group feedback, and 53.7% reported giving a combination of group and individual feed- back. Examples of behavioral checklists used in peer observa- tion can be found in Appendix C. Self-Management/Self-Observation Question 28 asked respondents three follow-up questions if they indicated using driver self-management/self-observation. One follow-up question asked respondents to indicate the frequency of the feedback given to drivers when using this BBS technique. Of the 21 respondents who indicated using self-management/self-observation, 19.1% gave daily feed- back, 14.3% weekly, 47.6% monthly, 9.5% quarterly, and 9.5% bi-annually (no respondents reported annual feed- back). Table 6 shows the effectiveness of self-management/ self-observation by the frequency of feedback as indicated by survey respondents. As can be seen in Table 6, effectiveness decreases as feedback becomes less frequent. However, because of the small sample size, these data should be interpreted with caution. Another follow-up question asked respondents to indicate how feedback is given to drivers. Of the 21 respondents who indicated using self-management/self-observation, 76.2% of respondents reported giving drivers feedback via a one- on-one meeting with the safety manager, 28.6% via private memo/letter, and 19.1% via public display, while 14.3% reported having drivers chart their own feedback. (The sum does not equal 100% because respondents could select more than one choice.) The last part of Question 28 asked respon- dents to indicate the type of feedback given to drivers. Of the 21 respondents who indicated using self-management/self- observation, 42.9% of respondents reported giving drivers individual feedback regarding their driving and non-driving behaviors, 9.5% reported giving group feedback, and 47.6% reported giving a combination of group and individual feed- back. Note that the percentage for individual feedback is low. As the premise behind self-management is that the individ- ual is self-accountable for recording and reviewing his/her own feedback, results for individual or combination of indi- vidual and group feedback should be close to 100%. These results suggest a deviation between practice and published BBS research. Examples of behavioral checklists that can be used to facilitate driver self-management can be found in Appendix C. Training and Education with Driving Behaviors Question 29 asked respondents one follow-up question if they indicated using training and education sessions on specific driving behaviors. This follow-up question asked respondents to indicate the frequency of the training and education sessions on specific driving behaviors. Of the 65 respondents (100%) who indicated using training and education sessions on specific driving behaviors, 10.8% held weekly sessions, 26.2% monthly, 33.8% quarterly, 20% bi- annually, and 9.2% annually (no respondent indicated daily training and education sessions). Table 7 shows the effective- ness of training and education sessions on specific driving Frequency of Feedback Daily Weekly Monthly Quarterly Bi-Annually Annually Effectiveness 2.75 2.3 2.5 2.0 1.0 * * No data points Number of respondents = 21 Table 6. Effectiveness of self-management/self-observation by frequency of feedback. Frequency of Feedback Daily Weekly Monthly Quarterly Bi-Annually Annually Effectiveness 2.7 3.0 2.6 3.0* 3.0 2.6 * Effectiveness rating based on one data point Number of respondents = 41 Table 5. Effectiveness of peer observation and feedback by frequency of feedback.

behaviors by the frequency of sessions as indicated by survey respondents. As can be seen in Table 7, effectiveness generally decreases as the training and education sessions become less frequent, with the exception of a rise for annual training. Examples of training and education programs can be found in Appendix D. Training and Education with Non-Driving Behaviors Question 30 asked respondents a follow-up question on frequency if they indicated using training and education ses- sions on specific non-driving behaviors. Of the 47 respon- dents who indicated using training and education sessions on specific non-driving behaviors, 8.5% held sessions weekly, 23.4% monthly, 40.4% quarterly, 14.9% bi-annually, and 12.8% annually. Table 8 shows the effectiveness of training and education sessions on specific non-driving behaviors by the frequency of sessions as indicated by survey respon- dents. As can be seen in Table 8, effectiveness slightly decreases as the training and education become less frequent (although the rating of effectiveness for annual training and educa- tion sessions was higher than bi-annual training and educa- tion sessions). Survey respondents rated training and education with specific driving behaviors more effective than with non- driving behaviors. Further, training and education sessions on specific driving behaviors were held more frequently than with non-driving behaviors. Incentives/Rewards Question 31 asked respondents two follow-up questions if they indicated using incentives/rewards with their drivers. One follow-up question asked respondents to indicate what types of rewards their organizations give to their drivers. Table 9 dis- plays the percentage of survey respondents who use each type of reward. (The sum does not equal 100% as respondents could select multiple rewards.) Most respondents indicated using some type of safety award (73.1%), such as a certificate or trophy to reward drivers, while few reported using paid leave (1.9%). Respondents were also asked to indicate what type of data is used to determine rewards. Table 10 displays the percent- age of respondents who use each type of data to determine rewards. (The sum does not equal 100% as respondents could make multiple selections.) Most respondents indicated using crash-free miles (86.5%) to reward drivers, while few respon- dents indicated using brake or speed data (7.7% and 9.6%, respectively). Respondents were also free to indicate selec- tions that were not included in the survey (i.e., “other”); these responses included rewarding drivers for injury-free days and attendance at safety meetings. As indicated previously, most BBS programs stress process-based incentives rather than outcome-based incentives. These data can be interpreted in several ways: (1) outcome-based data are the only type of data available to respondents, (2) respondents were unaware of the 30 Frequency of Sessions Daily Weekly Monthly Quarterly Bi-Annually Annually Effectiveness * 3.57 3.0 2.8 2.6 3.2 * No data points Number of respondents = 65 Table 7. Effectiveness of training and education session on specific driving behaviors by frequency of sessions. Frequency of Sessions Daily Weekly Monthly Quarterly Bi-Annually Annually Effectiveness * 2.8 2.9 2.6 2.2 2.5 * No data points Number of respondents = 47 Table 8. Effectiveness of training and education session on specific non-driving behaviors by frequency of sessions. Type of Reward Cash Paid Leave Private Recognition Safety Trinkets Public Recognition Tokens Safety Awards % Use (of 52 “yes” respondents) 57.7% 1.9% 30.8% 32.7% 63.5% 3.8% 73.1% Table 9. Use of rewards by survey respondents.

31 potential benefits of process-based measures, (3) outcome- based measures are much easier and less time consuming to assess than process-based measures, and (4) some combina- tion of the three. Disincentives/Penalties Question 32 asked respondents two follow-up questions if they indicated using disincentives/penalties with their drivers. One follow-up question asked respondents to indicate what type of penalty their organizations give to drivers. Table 11 displays the percentage of respondents who use each type of penalty. (The sum does not equal 100% as respondents could select multiple penalties.) Most respondents indicated placing a memo/letter in the driver’s file (87.7%) to punish drivers, while none (0%) of the survey respondents reported using public reprimands. Respondents were also asked to indicate what type of data they use to determine penalties. Table 12 displays the percent- age of respondents who use each type of penalty. (As previ- ously, the sum does not equal 100% as respondents could make multiple selections.) Most respondents indicated using crash data (93%) to punish drivers, while few respondents indicated using brake data (14%). Respondents were also free to state data sources that were not included in the survey (i.e., “other”). Responses (not shown in the table) included hours-of-service violations and idle time. Comparisons between data sources used for penalties (Table 12) to those for rewards (Table 10) are notable. Respon- dents were much more likely to use a variety of different data sources to assess penalties compared to rewards. Further, they were more willing to use performance data (i.e., speed and brake data) to punish drivers than to reward them. Again, fleet safety managers deviated from published BBS research and preferred outcome-based measures rather than process- based measures to assess penalties and rewards. Training/Coaching New Drivers Question 34 asked respondents to select five driving or non-driving behaviors they focus on the most when training/ coaching new drivers. Table 13 shows the percentages for the five most important driving and non-driving behaviors while training new drivers as indicated by survey respon- dents. (The sum does not equal 100% as survey respondents could select up to five driving and non-driving behaviors.) Most survey respondents focus on training new drivers how to conduct pre- and post-trip inspections of their vehicles (75.8%), while few focus on proper diet, exercise, or load securement (all 1.6%). Overall, most respondents focus their training with new drivers on driving behaviors rather than non-driving behaviors. Interestingly, two of the three most common driving behaviors addressed in these training and education sessions with new drivers (i.e., inattention and speeding) were also the two most frequent human behaviors found in all the crashes investigated at the three different levels in the Indiana Tri-Level Study (Treat et al., 1979). Inattention was also the most prevalent recognition error, and speeding was the second most prevalent decision error in the LTCCS. Type of Data Crash- Free Miles Incident- Free Miles Violation- Free Miles Customer/Public Comments Speed Data Brake Data Other % Use (out of 52) 86.5% 71.2% 44.2% 38.5% 9.6% 7.7% 11.5% Table 10. Data used by survey respondents to determine rewards. Type of Disincentives Memo/Letter in Driver’s File Points (Earned or Deducted) Private Reprimand Public Reprimand Cash % Use (out of 57) 87.7% 21.1% 86% 0% 14% Table 11. Use of penalties by survey respondents. Type of Data Crashes Incidents Violations Customer/Public Comments Speed Data Brake Data Other % Use (out of 57) 93.0% 87.7% 89.5% 65.0% 42.0% 14.0% 7.0% Table 12. Data used by survey respondents to determine penalties.

Training/Coaching Experienced Drivers Question 35 asked respondents to select five driving or non-driving behaviors they focus on the most when using refresher training/coaching with experienced drivers. Table 14 shows the percentages for the five most important driving and non-driving behaviors while refresher training/coaching experienced drivers. (Again the sum does not equal 100% as survey respondents could select up to five behaviors.) Most survey respondents focus on refresher training/coaching experienced drivers to drive attentively (70.5%), while few focus on diet, drugs, or load securement (all 1.6%). Overall, most respondents focus their refresher training/coaching with experienced drivers on driving behaviors rather than non- driving behaviors. In fact, this focus was more pronounced with experienced drivers compared to new drivers. As noted regarding training new drivers, the two most common driv- ing behaviors (inattention and speeding) addressed when training experienced drivers were also the two most frequent behaviors found in crashes investigated at the three different levels in the Indiana Tri-Level Study (Treat et al., 1979). Again, as noted previously, inattention was also the most prevalent recognition error, and speeding was the second most preva- lent decision error in the LTCCS. Part 4: Barriers/Problems to Implementing Behavioral Management Techniques The specific barriers/problems in implementing behav- ioral management techniques, mean ratings (to the nearest tenth), and rankings are presented in order in Table 15. When there were ties, rankings were determined by looking at additional decimal places. However, for simplicity, these are not shown in the table. The scale went from “0” (no barrier/ problem) to “4” (serious barrier/problem); thus, the higher the number, the greater the barrier/problem in implement- ing behavioral management techniques. The highest-rated barrier/problem was non-acceptance/lack of cooperation by drivers (mean rating of 1.7). The lowest-rated barrier/ problem was driver union (or other association) opposed to it (mean rating of 0.5). Most of the barriers/problems were rated very low, from slight barrier to neutral. Of course, given the sample was likely skewed toward managers who have actually implemented such programs, these data do not necessarily reflect the importance of these barriers for fleets in general. Part 5: Comments/Respondent Information In Part 5, a space was provided for respondents’ written comments. Approximately one-third of survey respondents made such comments. The comments focused on a variety of issues and expressed many different views. Some of the com- ments are provided below: • “After searching for some time for a BBS approach, we recently started piloting BBS using a driver self-inspection checklist at a few terminals. So far we are pleased with the results. Drivers participate in developing the checklist to cover key at-risk behaviors for their local operation. It is essentially a ‘post-trip’ checklist for their safety behaviors that day. We do not spend much effort on graphing the results for the terminal, use simplified Excel graphing for general trends by item at that terminal. We don’t spend any time graphing combined terminal results since all the 32 Safety-Critical Behaviors % Use (out of 62) Driving Behaviors Driving inattentively 54.8% Speeding 54.8% Following distances 54.8% Using mirrors 45.2% Lane changes 45.2% Driving fatigued 41.9% Backing maneuvers 37.1% Turning maneuvers 32.3% Identification of blind spots 12.9% Non-Driving Behaviors Pre- and post-trip inspection 75.8% Completing paperwork 25.8% Lifting techniques 12.9% Proper diet 1.6% Proper exercise 1.6% Other: load securement 1.6% Table 13. Driving and non-driving behaviors as the primary focus in training/coaching new drivers. Safety-Critical Behaviors % Use (out of 61) Driving Behaviors Driving inattentively 70.5% Speeding 59.0% Following distances 55.7% Driving fatigued 49.2% Lane changes 45.9% Using mirrors 34.4% Backing maneuvers 31.1% Turning maneuvers 26.2% Identification of blind spots 14.8% Non-Driving Behaviors Pre- and post-trip inspection 55.7% Completing paperwork 14.8% Lifting techniques 13.1% Proper exercise 3.3% Proper diet 1.6% Other: load securement 1.6% Other: drugs 1.6% Table 14. Driving and non-driving behaviors as the primary focus in refresher training/ coaching experienced drivers.

33 checklists are somewhat different. Employee support and ownership has been positive so far. Visible benefits are increased daily focus on safety behaviors, improvement through self-coaching, increased interaction among peers pointing out at-risk behaviors they observe co-workers committing and significant reduction of safety failures where we have it up and running.” • “The killer problem with BBS in the trucking environment is trying to find observation opportunities for lone work- ers many miles from the terminal working 24/7. We think the next best thing is to sell them why it helps them pro- tect themselves by reviewing their actions during the shift and recording the data they see. Thus, the self-observation checklists. We make a couple of assumptions about our co- workers in this process. They don’t want to experience safety failures any more than we do. They are essentially honest people who will admit they are human, record their self-observations and coach themselves to become the safest operator possible.” • “Drivers are lone workers. They represent a unique challenge in introducing and sustaining a BBS process. Empowerment and effective leadership are the keys to BBS.” • “I really like BBS, unfortunately there is not a whole lot out there that is easy and simple to implement. Our insurance carrier has a great one that I use a lot.” • “Our company is a very safety oriented company. Our safety program is supported strongly by upper manage- ment, and our drivers.” • “We have introduced Driver Scorecards which are basically a quality measurement of each driver. Training modules reduce the score while violations, accidents, motorist com- plaints, etc., add points. If drivers hit a certain threshold, we look at the points and determine if the driver needs to be terminated or if remedial training can salvage the driver. In addition, we’ve instituted a formal annual driver appraisal where we sit down with each driver and review their per- formance, establish training schedule, and solicit their feedback. We have elected to spend additional $ to retain drivers, not just the trinkets and picnics, but enhanced communication, driver involvement in problem solving, and some incentive programs (imagine an airline Frequent Flyer Program) where ‘points’ can be redeemed from a comprehensive catalog.” • “We currently have a BBS program and have been success- ful with it.” • “BBS can only be successful with management buy-in.” • “We currently have a BBS program and have been success- ful with it.” • “Began BBS a year ago. Still shaking down the BBS sys- tem and implementation with personnel in the organi- zation due to new software, new incident investigation forms and techniques, and observation and feedback are all new.” • “. . . everyone makes BBS decisions all the time . . . with- out awareness and techniques training we miss opportu- nities to help ‘shape’ the BBS decisions toward safe behavior.” Survey respondents were also asked to provide general demographic information about themselves and their fleets. The 65 fleet safety manager respondents had been safety managers at their current fleet for an average of 11.7 years (range: 1 to 42) and had an average of 23.2 total years of experience in CMV operations (range: 1 to 44). Fleet size varied widely, ranging from 2 to 7,500 power units. The median fleet size was 101 power units. Respondents were also asked to characterize their fleet’s primary operation by selecting one of seven major truck and/or bus operation types or writing in an alternative. Results are shown in Table 16. (The sum is more than 100% as some fleets had more than one operation type.) As can be seen in Table 16, private fleets were likely over- represented in the sample. Thus, readers should be mindful of this over-representation when interpreting the results pre- sented in this synthesis. One possible interpretation of this over-representation is that private fleets are more likely to Fleet Safety Managers Barriers/Problems Mean Rank Non-acceptance/lack of cooperation by drivers 1.7 1 I and/or other company safety managers don’t know enough about it 1.2 2 BBS takes too much time to implement 1.1 3 Not enough money/budget to support it 1.1 4 We tried BBS techniques, but they have not worked well 0.9 5 Company top management doesn’t support it 0.8 6 Driver union (or other association) opposed to it 0.5 7 Number of respondents = 65 Table 15. Barriers/problems in implementing behavioral management techniques.

implement BBS program (no data could be found to support this claim); however, it’s also possible these types of CMV operations were over-sampled in the synthesis. Key Findings • Few focus group and survey respondents have imple- mented a comprehensive BBS program; rather they have implemented behavioral management techniques in a piecemeal fashion. • The highest-rated driving behaviors (tailgating, speeding, and speeding on curves) with respect to crash risk were similar to those identified in major crash database studies (e.g., LTCCS and Indiana Tri-Level study). • More than one-third (35.4%) of survey respondents use some type of OBSM device to observe driver behaviors. • Almost all survey respondents (83.1%) use some type of observation approach to observe driver behaviors. • All respondents used some type of behavioral management approach to improve the safety of their drivers. • Training and education with respect to driving behaviors was the highest-rated behavioral management technique (2.9 out of 4.0). • There were several discrepancies between research (the BBS research literature) and practice (as indicated by sur- vey respondents), including – Frequency of feedback and – Type of data used to determine incentives and penalties. • Training and education of new drivers focus primarily on pre- and post-trip inspection and several driving behaviors (driving inattentively, speeding, and following too close). • Training and education programs with experienced drivers, also called refresher training, focus mostly on critical driv- ing behaviors (driving inattentively, speeding, and following too close). • Respondents do not view the barrier/problems to imple- menting behavioral management techniques as significant impediments, perhaps because all survey respondents have implemented some type of behavioral management tech- niques with their drivers. 34 Operation Type # Fleet Safety Managers % Fleet Safety Managers For hire: long haul/truckload. 12 18.5% For hire: long haul/less-than-truckload (LTL) 4 6.2% For hire: local/short haul (most trips < 100 miles) 5 7.7% Private industry: long haul 16 24.6% Private industry: local/short haul (< 100 miles) 22 33.8% Passenger carrier: long haul/motor coach 6 9.2% Passenger carrier: local/transit 3 4.6% Table 16. Safety managers’ fleet operation types.