Identifying and Reducing Workforce Fatigue in Rapid Renewal Projects (2014)

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.

46 C h a p t e r 6 An integrated approach to fatigue risk management for high- way construction work needs to accommodate the unique char- acteristics of the environment, including the general industry approach to safety, the seasonal nature of construction work, and the likelihood of unpredictable schedule changes due to various factors such as weather, unforeseen obstacles, re-work, and so forth. This chapter discusses an overall organizational approach to fatigue risk management, including processes and implementation steps for organizational practices that are gen- erally applicable across contractors of various sizes, and work schedule and work practice guidance based on fatigue model- ing of schedules typically encountered in rapid renewal con- struction. Fatigue management is a joint responsibility between management and individual employees; this section establishes processes to implement the collaborative approach. Organizational practices Guidance This section describes adaptations of organizational practices for fatigue risk management that are appropriate for a self- regulated industry such as highway construction, with a focus on monitoring and mitigation. The approaches described in this section are meant to be flexible and adaptive, so that they can apply to a broad range of organizational size and complexity. Figure 6.1 illustrates the elements of organizational practice to address and implement fatigue risk management in highway construction firms (i.e., contractors). These practices would also be applicable to state employees if they are not already covered by work-hour limitations in labor agreements. The upper part of the figure lists general processes for fatigue risk management, and the bottom part of the figure lists specific implementation means to institutionalize those processes. The following sections discuss each of these process and implemen- tation steps. Assess Corporate Approach The first process step in addressing fatigue risk management in highway construction is identifying the current corpo- rate approach. The most fundamental question is whether an approach to fatigue management exists. The team’s field sur- vey work suggested that managers believed their corporate safety training addressed issues of fatigue, but the team did not see any substantiating material. Instead, it seems that fatigue issues, if they are addressed at all, revolve around proper hydration and physical rest breaks in extremely hot weather, leaving fatigue from sleep loss and circadian rhythm misalignment (night work) unaddressed. An enabling process for fatigue risk management is a corpo- rate Safety Management System (SMS). It is likely that national- level construction firms have such systems in place, whereas smaller and regionalized firms may have more informal approaches. In either case, it is important to assess the extent to which fatigue risk management is or is not addressed. If an SMS exists, it can be reviewed for any mention of fatigue risks and mitigations and for appropriate places in which manage- ment processes might be inserted to address the problem. Existing safety processes that may be adapted and extended to the worker fatigue problem include incident investigation and reporting and worker input procedures. While safety tends to be viewed as a shared responsibility between management and staff in organizations, there are certain roles and responsibilities in construction firms that will likely have a closer connection with worker fatigue than others. Based on the team’s field research these staff roles appear to be superintendents, construction engineering plan- ners, and labor crew supervisors. Planners have a key role in establishing specific construction tasks to be carried out, and this interacts with when they would be carried out, including the potential need for closures and night work. Superinten- dents tend to be aware of the pace and intensity of work and Organizational Approach to Fatigue Risk Management in Rapid Renewal Highway Construction

47 the likely productivity and safety impacts on workers. Crew supervisors would have a similar and more immediate under- standing of fatigue issues on specific crew members. Build-in Fatigue Management Assessment of the basic corporate approach to safety should lead to a concrete implementation step of building fatigue management into the overall process. The initial requirement for this step is management concurrence. As in virtually all cor- porate initiatives, leadership commitment is essential, not only for approving whatever resources may be necessary (and this may be a relatively small amount of personnel time in most cases), but also for reinforcing messages and business practices. Changes in scheduling may impact overall performance, so fatigue mitigation should be considered by executives. Specific methods for incorporating fatigue management in the corporate safety approach include obtaining and analyz- ing data and enhancing safety training. In terms of data, information that reflects on the extent to which fatigue may be a problem is important. This may be as simple as repeated ver- bal reports from personnel concerning scheduling issues or more detailed data reflecting productivity or safety incidents on different shifts. There will be wide variation across organi- zations in the nature of the data, and how they are obtained and analyzed, depending on the size and complexity of the contracting firm. Safety training is one of the first lines of defense in fatigue management, and the team’s field research suggests that exist- ing training covers this topic only incidentally, if at all. While there are different models for training, such as new employee orientation and safety training, project-specific training, and daily crew briefings, each of these provides an opportunity to incorporate information about fatigue management. The Guide to Identifying and Reducing Workforce Fatigue in Rapid Renewal Projects contains an extensive compilation of train- ing materials appropriate for both workers and management which can be adapted to specific organizational approaches. Dispel Erroneous Beliefs The field research identified a number of inaccurate attitudes and beliefs held by a considerable percentage of workers, and a lower percentage of management, concerning fatigue and how to deal with it. These beliefs tend to transform into “myths” over time, influencing how people think and com- municate about fatigue, regardless of accuracy. Some of these inaccurate beliefs include • Fatigue is something to muscle through; • Fatigue management is a personal responsibility; • Fatigue is inevitable; • Napping is not okay in the work place; and • Everyone has enough time off for recovery. Figure 6.1. Organizational practices for implementing fatigue risk management in highway construction firms. Assess Corporate Approach • Safety Management System • Roles and responsibilities • Existing Safety Processes Build-in Fatigue Management • Get Leadership commitment • Obtain and analyze data • Enhance safety training Dispel Erroneous Beliefs • Implement science- based training • Treat fatigue as a safety problem • Encourage Culture Change Analyze Fatigue Risk Trajectory • Performance errors • Sleep obtained • Sleep opportunity Assess Schedule Risks • Use model-based analysis • Determine relative risks • Identify counter measures Formalize Risk Assessment Process • Conduct for each project and phase • Use as a basis for scheduling workers • Use as a basis for construction planning and project bidding Implement Counter measures • Training for all workers • Strategic napping periods • Fatigue proofing • Caffeine strategies Reporting, Investigation, Evaluation • Evaluate modified schedules • Track and investigate incidents • Encourage worker reporting Timeline for Implementation

48 To the extent that beliefs such as these prevail and are per- petuated as myths through various elements of the work- force, fatigue will not be treated seriously. Thus, an important continuing process is to gradually dispel these beliefs and alter the cultural view of fatigue. Science-based training can form the basis for addressing erroneous beliefs, through use of such material as provided in the Guide to Identifying and Reducing Workforce Fatigue in Rapid Renewal Projects. Training can help to shift the concep- tion of fatigue from that of “inevitable annoyance” to that of “safety problem” and establish a basis for cultural change to seriously address it through fatigue risk management. Analyze Fatigue Risk Trajectory Implementing a process to dispel erroneous beliefs can be facil- itated by using an analytic framework to clearly link safety issues in highway construction jobs with fatigue as a causal factor. The fatigue risk trajectory (Figure 6.2), based on research by Dawson and McCulloch (2005), provides a means for understanding the pathways to safety problems that can be used by safety managers for initial job/task and schedule analy- sis. The basic trajectory involves opportunities for sleep pro- vided by work schedules, sleep obtained, on-the-job fatigue, and fatigue-related errors. Behavioral outcomes and counter- measures associated with these risk factors can be used as a basis for intervention. An illustration of how the fatigue risk trajectory can be used to analyze potential error-prone situations is shown in Figure 6.3 (adapted from Reason’s Swiss cheese model of accident causation; Reason 1990). Errors occur when “holes” in the defensive layers align and prevailing circumstances enhance their likelihood. An example would be a maintenance- of-traffic worker scheduled on successive 12-h night shifts, and getting less than 5 h of sleep during each off period. That individual is fatigued cumulatively throughout the week and may fail to implement traffic routing procedures correctly. Due to schedule pressures or unavailability of personnel to verify the placement of traffic routing diversions or conditions such as poor visibility or rain, this is not noticed and the result is vehicle incursion into the work zone with resulting injuries. Similar problems could occur with setting up con- struction equipment or rigging, putting multiple personnel at risk. Critical in this perspective on error causation is that multiple problems line up to cause an incident or accident, and fatigue is often one of those problems and therefore a risk factor to be managed and mitigated (Van Dongen and Hursh 2010). Figure 6.2. Fatigue risk trajectory. Risk Factor Error Trajectory Focused Countermeasure Sleep obtained Individual choice of time use Training and strategies to optimize available break time for recovery Fatigue-related errors Task performance lapses, injuries, accidents Work zone safety practices, oversight procedures Sleep opportunity and Circadian factors On-the-job fatigue Insuf f icient break length Behavioral symptoms, inattention, poor decision- making Revised scheduling Symptom checklists, supervisor and peer observation, rest/nap breaks, appropriate use of caf feine Adapted from Dawson and McCulloch 2005.

49 Assess Schedule Risks A systematic approach to risk assessment can utilize knowledge of how fatigue occurs over the course of work periods for staff on various schedules. This approach can be facilitated with a computer-based model, although for most construction firms some heuristics based on model outputs contained in the Work Scheduling Guidance will likely be sufficient. A method for modeling worker fatigue levels associated with various sched- ules is described in the next section of this report. The models can be used to determine likely fatigue levels for workers, based on the schedules they are assigned and how long they have been on them. This information can be used to evaluate the recovery opportunities provided by exist- ing and planned worker scheduling. Construction planning for specific skills and crafts across the 24-h period in different phases of projects will influence worker scheduling. Planners should evaluate the impact of construction scheduling require- ments in terms of worker fatigue impacts and try to ensure that work schedules dictated by construction requirements do not adversely affect individuals or groups of workers. The models can also show when commuting is likely to be a safety issue, such as the night shift, when driving home occurs at the peak fatigue level. Fatigue profiles such as this are also useful for evaluating napping opportunities when they might occur in the work shift and their impact on fatigue levels. For example, if there is a desire to reduce peak fatigue prior to driving home after a night shift, fatigue profiles can be used to show the increase of fatigue throughout the night and provide comparative pro- files with and without naps. Similarly, use of earlier work stop times on shorter night shifts can be compared with later start times to show daily peak fatigue levels and also accumulation throughout the week. Finally, it is important to address the work hours of design- ers and managers, especially as they work night shifts follow- ing day shifts, or participate in long closures followed by a full week of day-shift work. Formalize Risk Assessment Process Implementation of the risk assessment process should eventu- ally be undertaken as a regular activity, starting with analysis of contract opportunities and continuing through the bid, con- struction planning, and execution of each project phase. Since rapid renewal projects often involve alteration of construc- tion activities due to emergent circumstances, any schedule revisions should be reviewed as well. For example, scheduling of crews involves interaction between construction engineer- ing and crew superintendents, and to the extent that superin- tendents see certain work crews affected by, say, too much night work, they should negotiate the execution of various construction tasks so that the crew are provided with recovery opportunities. This may involve work breaks, naps at the work site during night shifts, re-scheduling certain tasks for day work if possible, increasing staffing, and generally providing relief from constant night work. Fatigue risk assessment can also be used as a formal process for construction planning, as well as for determining schedule and fatigue impacts of projects in the bid evaluation stage. For example, if a request-for-proposal contains incentives for completion or a specific number of closures permitted, mod- eling could be used to determine the work schedules required, availability of crew for such schedules, and potentially whether Figure 6.3. Fatigue risk trajectory and multiple levels of defense. Adapted from Reason 1990. Injury or Fatality Sleep Obtained Fatigued Personnel Procedures Physical Conditions Work Schedule Risk Factors

50 the project warrants bidding. If schedule modeling were to show night work at a level that management considers unsus- tainable, alternative approaches might be proposed. Implement Countermeasures Countermeasures for fatigue are an important component of an overall organizational approach. The primary countermea- sure is education and awareness for all personnel, including management, to dispel the myths and erroneous beliefs about fatigue, and to instill an understanding of the biological basis of fatigue and the things that can be done about it. A few key countermeasures have been found to be effective in a variety of industrial environments, including defensive napping prior to night work and napping at appropriate times during the work period (such as the lunch break), caffeine during periods of high fatigue or to reduce sleep inertia (the fatigued feeling upon waking) after mid-shift naps, rest breaks from the work flow, and using scheduling to try to accommo- date individuals who have varying susceptibility to fatigue. One potential approach to a work break is to consume a caffeinated beverage just before a 30-min nap, and at the end of the nap the caffeine will be starting to take effect. If the caf- feinated beverage is cold rather than hot, this may facilitate rapid consumption if time for the nap is limited. This approach will have the dual impact of reducing sleep inertia and reduc- ing fatigue for the following several-hour period. The team suggests implementing countermeasures as a rela- tively continuous process, rather than a discrete implementa- tion step. This is because conditions in rapid renewal projects are dynamic, and the specific approaches to implementation may vary with the schedule and season. For example, night work might be scheduled for a somewhat earlier start in the summer months, leading to a work stop time that allows work- ers to get home and into bed before it is completely light out- side. It has been reported in the team’s field work that this facilitates getting to sleep faster and sleeping somewhat longer, and this observation is supported by circadian physiology. Fatigue countermeasures involve not only mitigating fatigue through rest breaks, better sleep opportunities, and so forth, but also addressing the fact that fatiguing schedules cannot be entirely eliminated. Night work is a fact of life in rapid renewal highway construction. In addition to addressing fatigue- reducing countermeasures, there are fatigue-proofing strategies for adding layers of defense against error. These include • Increased supervisory oversight; • Use of written procedures and checklists; • Self- and peer-monitoring during critical periods; • Reducing monotonous or highly complex tasks during periods of high fatigue; • Extra personnel for critical and dangerous tasks; • Nap timing for best impact; • Interaction with peers to evaluate fatigue levels; • Self-selected rest breaks; • Transportation assistance following extended shifts; and • Training for workers and managers in how to recognize fatigue. By continually evaluating schedules and conditions of work, safety managers can adapt both fatigue-mitigating and fatigue- proofing countermeasures to prevailing conditions. Reporting, Investigation, Evaluation The role of fatigue in construction safety problems is probably under-represented due to lack of reporting and investigation. A proactive management approach to fatigue should encour- age workers to report problems, whether they are related to scheduling, specific tasks, or even other workers. In order to better understand specific project fatigue prob- lems, safety incidents should be investigated with the fatigue factor in mind, including whether night work was involved and individuals worked many successive night shifts without a break, whether a weekend closure was involved, and whether the individual workers were experiencing sleep restriction or sleep problems. Commuting accidents, although technically not occurring during duty hours, can sometimes be related to fatigue from night work, and are especially likely during rush traffic in the morning. Information collected can be useful in modifying work schedules. A primary issue in implementing this step relates to the availability of data, determining what the current pro- cedures are, if any, to document and investigate incidents and accidents. This will vary considerably across organizations based on their size and complexity. Trade and government organizations may play a role in providing standards and tools for structured data collection efforts. The role of sleep disorders in contributing to workplace fatigue should not be ignored. In the field survey, eight (17%) out of 47 survey respondents reported having been diagnosed with a sleep disorder in the past, but only half reported receiv- ing treatment. Sleep disorders are associated with excessive daytime sleepiness and increased prevalence of motor vehicle accidents and occupational injuries (Findley et al. 1988; Aldrich 1989; Chau et al. 2004; Chau et al. 2004). The majority of shift workers have a sleep disorder (Leger 1994), and some sleep disorders can be caused by shift work (Guilleminault et al. 1982). Common sleep disorders include insomnia, restless leg syndrome, and obstructive sleep apnea. Obstructive sleep apnea is particularly prevalent among men between the ages

51 of 30 and 60, a risk that increases if they are overweight (Bixler et al. 1998). Evaluation and treatment of sleep disorders has benefited the trucking industry. Summary of Organizational Practices Figure 6.4 shows how to integrate the organizational practices described in this section. An overarching safety management system can be anything from informal processes conducted by an individual part-time safety officer in a small firm, to a more formalized structure with procedural mechanisms and formal reporting documentation and channels in larger organizations. The fundamental outputs of the organizational practices are the same: a deliberate and rational method for addressing and mitigating the impacts of fatigue on opera- tional personnel. training Material Two training modules were developed: a basic fatigue train- ing program intended for all workers and a training program intended for managers and other supervisory personnel. The general objectives of training include the following: • Raise awareness of the fatigue issue, particularly as it per- tains to rapid renewal work schedules. • Establish a knowledge base for understanding and respond- ing appropriately to fatigue issues. • Provide workers and managers with specific strategies and tools for avoiding fatigue or reducing and mitigating the effects of fatigue on the job. Detailed training material content is contained in the Guide to Identifying and Reducing Workforce Fatigue in Rapid Renewal Projects. Basic Training The Basic Fatigue Training is intended for use by all persons involved in highway construction, including laborers, super- visors, managers, designers, and DOT personnel. Ideally, it should be completed by managers and other relevant person- nel before completing the manager training. Basic training provides an introduction to fatigue concepts (e.g., circadian cycle and sleep debt), explains the causes and consequences of fatigue, and outlines practical strategies (individual counter- measures) for preventing, recognizing, and managing fatigue at work. The training begins with a set of learning goals, and each module concludes with questions for discussion or per- sonal consideration, intended to encourage workers to evalu- ate the concepts in light of their own experiences and to apply strategies based on their own needs. Information is tailored specifically to the demands of the highway construction envi- ronment; however, the training content is also broadly appli- cable and will be relevant to workers in organizations with different roles and contexts. Adapted from Gander et al. 2011. Safety Management System • Based on culture • Collaborative • Proactive Fatigue Risk Management System Policy Steering Committee Education and Training Reporting Inputs • Incidents/Investigations • Voluntary fatigue reports Outputs • Work schedules for specific project phases • Personnel schedule rotations • Counter measure plans and implementation • Contract bidding and staffing strategies Analysis Inputs • Construction schedule requirements • Fatigue impacts of schedules • Counter measure effectiveness Figure 6.4. Integrated elements of fatigue risk management system.

52 Manager Training Manager training summarizes some of the basic fatigue material but is meant to provide a larger organizational con- text, processes, and implementation steps for addressing fatigue risk management. The focus is on schedule risk assess- ment in terms of sleep disruption and how various schedules are used in different phases of project execution. Illustrations of fatigue impacts based on work schedule models are pro- vided, and there is considerable discussion of the relationship between safety management as it is currently implemented and how fatigue risk management might fit in. The training is meant to convey an understanding of the basic processes and steps of fatigue management, and it encourages partici- pants to evaluate their own organizations to determine how these approaches might be implemented. There is wide varia- tion in construction company size and complexity but fatigue management is an important issue across these variable con- texts. One approach will not fit all circumstances, but the training conveys a series of analytic questions that will allow participants to evaluate their unique organizations and adapt fatigue risk management methods accordingly. Work Scheduling aids and Work practice Guidance Work scheduling and work practices guidance was developed to address worker fatigue associated with different schedules used in rapid renewal highway construction. The basic form of this guidance consists of schedules illustrating employees’ work start–stop times and sleep–wake times for each day of the week, fatigue profiles over the 24-h period for the week- long schedule, and recommended fatigue mitigations for each schedule. In this section, the team first discusses the basic variations of rapid renewal scheduling and the factors that influence work schedules, then provides a discussion of a sample of the work schedule and work practices guidance contained in the Guide to Identifying and Reducing Workforce Fatigue in Rapid Renewal Projects. Rapid Renewal Schedules and Implementation Considerations The most common types of work-hour scheduling for high- way construction contractors include • Daytime construction; • Nighttime construction; • Continuous construction; and • Combinations of the above. Weekday daytime construction generally does not involve major lane closures for urban highways in order to avoid congestion associated with a traffic shift. In this case, the con- tractor maintains its construction work behind K-rail with traffic pushed to the side. In rural areas, daytime construction can be performed with a lane closure in place, with the con- tractor performing work behind traffic barriers (such as rub- ber cones or plastic barrels), as long as the lane closure does not cause major traffic congestion. Nighttime construction typically involves some degree of lane closure, with short closures of 5 to 7 h for highly con- gested urban areas, longer closures of 8 or more hours for urban areas with less traffic density, and extended closures of up to 11 h for rural locations. Continuous construction involves major lane closures in urban areas to perform large-scale renewal work. The practical implementation of work scheduling by con- tractors for specific projects depends on the lane closure guidelines and requirements specified in the sponsoring agency’s transportation management plans, or management of ongoing traffic. In general, a state DOT develops lane clo- sure charts (also called “lane open charts” or “lane require- ment charts”) for a highway renewal project as part of the transportation management plan. The lane closure charts dictate how many lanes can be closed for construction during specified hours (alternatively, the lane open charts show how many lanes are required to be open for each hour). In addi- tion to these constraints, contractor work-hour schedules are influenced by (1) working days estimated by the state DOT for the bid, (2) Cost + Schedule (so-called “A + B”) contract elements, and (3) incentives and disincentives in the contract. Additionally, the type of work to be performed at various phases of construction (such as bridge or utility work) is combined with these three primary project management elements. Work Schedule Fatigue Modeling A critical piece of information for managing worker fatigue is understanding the combined influence of biological rhythms, work schedule, and sleep patterns on fatigue during the work period. By understanding these variations, managers and engi- neers can address problems through various countermeasures and interventions to mitigate fatigue. The team developed fatigue profiles for 20 basic schedules (5 day schedules; 5 nights; 4 closure weekends; 4 switching shifts; 2 manager/designer) that may be encountered in rapid renewal construction, with variations based on applications of countermeasures, for a total of 107 models. This section discusses the technical basis for these fatigue model profiles, illustrates how they can be applied, and discusses the general categories of risk factors and countermeasures (work prac- tices) that may be applied.

53 Fatigue Modeling Technical Basis Based on a recent literature review, Dawson and McCulloch (2005) determined that work schedules resulting in the fol- lowing conditions would be inconsistent with safety in the workplace: • Less than 5 h of sleep in the prior 24 h; • Less than 12 h of sleep in the prior 48 h; or • Longer wakefulness than the total amount of sleep obtained in the prior 48 h. However, this rule of thumb depends critically on prior circumstances, most notably on the circadian timing of the work period. A more scientifically valid and operationally optimal approach makes use of a mathematical model of fatigue to forecast the fatiguing effects of a work schedule (Raslear et al. 2011). This is the basis of model-based fatigue risk management (Van Dongen and Belenky 2012), which is gradually gaining widespread acceptance in U.S. operational settings. The model used for the illustrations in this report is an expansion of a biomathematical model of fatigue developed by McCauley et al. (2009). This model is the first to incorpo- rate the neurobiology of long-term changes in the homeo- static equilibrium for sleep/wake regulation—in other words, changes in sensitivity to future sleep loss due to exposure to sleep loss in the recent past. In the expanded version of the model used here (McCauley and Van Dongen 2012), the cir- cadian rhythm in the effects of fatigue interacts dynamically with the homeostatic changes to accurately predict the fatigu- ing effects of sleep loss in night work operations. The model is calibrated to predict lapses on the psychomotor vigilance test (PVT) (Lim and Dinges, 2008), a gold standard measure of fatigue, based on three large data sets from published labo- ratory studies of sleep loss and circadian misalignment and validated using another three such data sets. The resulting profiles may be considered to reflect the overall rise and fall of an underlying construct, fatigue, which is correlated with more performance errors, reduced subjective alertness, and increased likelihood of falling asleep. It is tempting to think of fatigue model output as defining hard constraints concerning when a person is sufficiently rested versus too fatigued to work. However, fatigue models are not meant to be employed in this way for several reasons. First, while there is an association between fatigue and safety incidents, the existence of fatigue does not always lead to acci- dents; a fatigue-related error must coincide with a safety-critical condition to result in an accident. Second, there are individual differences in susceptibility to fatigue and performance effects; models provide results for a statistical average. Finally, the risk associated with a given level of fatigue depends also on the task at hand, the operational circumstances, and the tolerance level for performance impairment and errors (see Figure 6.5). The model outputs are meant to contrast fatigue in terms of relative risk, that is, the likelihood that various schedules will Original graphic created by H. P. A. Van Dongen June 2012. Work and Social Hours (Duration) Work and Social Hours (Timing) Operational Environment (effects attributable to the job) sleep inertia Human Neurobiology (effects attributable to the individual) circadian drive for wakefulness homeostatic drive for sleep cognitive functioning cognitive performance time on task → Relative Risk of Errors, Incidents, Accidents instability trans ient effec t transient effect Other Risk Factors (hazards, work demands, time pressure, distractions, weather, etc.) Counter- measures (caffeine, rest breaks, automation, extra staff) C O N STR U C TIO N SC H ED U LIN G C O N STR A IN TS Light Level Figure 6.5. Neurobiological and operational determinants of fatigue illustrating interaction between physiological and work factors, relative risks, and performance outcomes.

54 induce greater or lesser degrees of fatigue based on the inter- action of multiple work schedule factors. Fatigue models provide a guide to schedules and times of day that are more likely to be associated with fatigue and safety problems. However, like traffic and weather forecasts, fatigue models are not predictive for any one individual, schedule, or project in an absolute sense. Whereas models do not (and, by definition, cannot) provide strict assessments of whether a given schedule is safe or unsafe, they are very useful as tools to evaluate planned schedule improvements that will help reduce fatigue, suggest good practices for administration of fatigue countermeasures, and guide decisions about appro- priate tasking and level of supervision based on the relative fatigue likelihood associated with a schedule. Table 6.1 shows the basic schedule combinations the team modeled for generic shift scenarios. The scenario combinations were reported during the field research. The main purpose of the schedule modeling was to assess variations in fatigue that are manifest from fundamental changes in schedule, such as day versus night work, as well as from more subtle influences such as the impact of consecutive days or nights of work, strategic napping, and sleep–wake schedule maintenance during days off. The team made several assumptions when modeling schedules: • All schedules include a 1-h commute each way and a 30-min meal break. • Commute time is not included in work or sleep time, but meal breaks are included in work time (i.e., time “on duty”) such that a “10-h” shift takes place over 10 h, 30 min. • Day shift schedules allow 30 min in the morning for per- sonal tasks between waking up and leaving for work. Table 6.1. Scenarios, Assumptions, and Variables Used in Work Schedule Fatigue Modeling Scenario Assumptions Schedule and Countermeasure Variables Modeled Day Shifts • 30-min meal break • 1-h commute • 7.5 h sleep nightly • 7:00 a.m. start time • Length of day (7, 8, 10, 12 h) • Length of week (40, 48, 50, 55, 60 h) • Alternating Saturday work Night Shifts: 5, 7, 8, 11 h closures • 30-min meal break • 1-h commute • 1:00 p.m. wake time • Length of day (8, 10, 11, 12 h) • Length of week (40, 48, 50, 55, 60 h) • Naps at work • Defensive nap at home before night shift • Reversion to day schedule on days off vs. maintenance of night schedule Weekend Closure: 55 h • 30-min meal break • 1-h commute • 1:00 p.m. wake time when on nights • 7.5 h sleep nightly when on days • 7:00 a.m. start time when on days • “Standard” 5 × 10 schedule during week with 12-h shifts on weekend • All day shifts • All night shifts • Day shift during week switching to nights over weekend closure • 12 days straight vs. 1 day off • Naps at work when on nights • Defensive nap at home before night shift Switching Shifts • 30-min meal break • 1-h commute • 1:00 p.m. wake time when on nights • 7.5 h sleep nightly when on days • 7:00 a.m. start time when on days • 10-h shifts • Days to nights over weekend off • Days to nights midweek with day off • Days to nights midweek without day off • Nights to days over weekend off • Nights to days midweek with day off • Nights to days midweek without day off • Naps at work when on nights • Defensive nap at home before night shift Manager and Designer • 30-min meal break • 1-h commute • 7:00 a.m. start time • Manager 55-h week, 7.5 h sleep nightly • Designer 50-h week, 7.5 h sleep nightly • Designer 80-h week, 6 h sleep nightly Anchor Sleep Schedules • 30-min meal break • 1-h commute • Day shift • 6.5 h sleep nightly for laborer • Manager weekend closure • Laborer, 6:00 a.m. start (5 × 10) with after-work nap • Laborer, 5 × 12 day shift with nap at work Restricted Sleep • 30-min meal break • 1-h commute • 5 × 10 day shift • 7.5 h sleep nightly when not disrupted • 1 night of 4.5 h sleep • 2 consecutive nights of 4.5 h sleep

55 • Night-shift schedules allow 1 hour in the morning between arriving home and going to bed, for personal tasks and “winding down” after work. • Personal task periods are not included in sleep time or work time. • Day shift schedules assume 7.5 h of sleep per night, unless otherwise noted, regardless of shift duration. • Night-shift schedules assume a wake time of 1:00 p.m. between night shifts due to circadian pressure to wake at this time; hours of sleep obtained are therefore determined by the end of shift. • The final night shift in a week is followed by a morning nap of 2 h if reverting to a day schedule on days off. Detailed modeling results are contained in Appendix D. The following section summarizes the results from the schedule modeling. Illustrative Model Results Figure 6.6 through Figure 6.8 show sample results from mod- els of day and night shifts, with a work week consisting of five Figure 6.6. Fatigue profile: 5  10 (50-h week) day shift. Figure 6.7. Fatigue profile: 5  10 (50-h week) night shift with mid-shift and defensive naps. Figure 6.8. Fatigue profile: 5  10 (50-h week) night shift without naps.

56 10-h days, Monday through Friday. Work start and stop times are illustrated, as is the sleep period. These shifts were chosen because they illustrate common phenomena that occur across schedules. Three 50-h week shift variations are depicted: 5 × 10 day shift (Figure 6.6); 5 × 10 night shift ending at 6:30 a.m., with mid-shift naps and defensive naps (Figure 6.7); and 5 × 10 night shift ending at 6:30 a.m. with no naps (Figure 6.8). Note that fatigue plots and bar graphs in this chapter employ consis- tent vertical scales. Initial inspection of the model output shows that there is a substantial difference in overall fatigue levels between day and night shifts. While there is variation in fatigue level in the day shift (Figure 6.6), both the peaks and troughs for the night shift are substantially higher than for day shift, and the difference between the peaks and troughs is much greater on night shift. The day shift shows a rise in fatigue that peaks just after noon, with a decline toward early evening followed by a sharp rise just before bedtime, and no accumulation of fatigue throughout the week (Figure 6.6). On the night shifts (Figure 6.7 and Figure 6.8), fatigue rises throughout the work period and the commute home and peaks at bedtime. Part of this effect is accounted for by the lower amount of sleep typically obtained by night shift workers—5 h, rather than 8 h (Åkerstedt 2003)—due to dif- ficulty sleeping during that portion of the circadian cycle. The overall effect of this is cumulative; fatigue is higher on successive days of the week, reaching its maximum at bedtime following the last 10-h shift. A mid-shift nap of 30 min each night and a defensive nap the afternoon before the first night shift reduces peak fatigue each night and the cumulative effect throughout the week (Figure 6.7), relative to no naps (Figure 6.8). Figure 6.9 compares the peak fatigue reached after start of the work period for each of the schedule models shown in Fig- ure 6.6 through Figure 6.8. The differences in peak fatigue and cumulative effects between schedules are more evident here. Figure 6.10 through Figure 6.12 contrast day shift fatigue profiles under three different conditions: normal sleep of 7.5 h and 1 or 2 nights of restricted sleep (4.5 h each night). The restricted sleep occurs on Sunday (Figure 6.11 and Figure 6.12) and Monday nights (Figure 6.12 only), and resulting higher levels of fatigue week relative to the profile with no restricted sleep (Figure 6.10) can be seen on Monday and throughout the rest of the week. The impact of 2 nights of restricted sleep leads to the highest levels of fatigue on subsequent days, and the effect lasts longer (see Figure 6.13 for a comparison of peak fatigue levels). It takes more than a single good sleep period to recover from 1 night of serious sleep deprivation, and more still for 2 nights. These results suggest that project managers should pay close attention to the potential longer-term impacts of short-term sleep restriction. Modeling Summary and Work Practice Guidance Implications This section discusses the work schedule modeling findings that are the most significant for influencing work practices such as scheduling decisions, shift start and stop times, closure lengths for various construction phases, and specific fatigue countermeasure implementation within a schedule. Daytime Construction Daytime construction schedules are preferable as a means of minimizing fatigue and obtaining adequate recovery sleep. Monday Tuesday Wednesday Thursday Friday G re at er F at ig ue day shift night shift, with naps night shift, no naps Figure 6.9. Peak fatigue: 5  10 (50-h week) day shift and night shift with and without naps.

57 Figure 6.10. Fatigue profile: day shift with normal sleep (7.5 h per night). Figure 6.11. Fatigue profile: day shift with 1 night restricted sleep (4.5 h). Figure 6.12. Fatigue profile: day shift with 2 nights restricted sleep (4.5 h).

58 Models of daytime construction schedules (40 to 60 h per week) show no differences in fatigue profiles across shift type or throughout the week, assuming 7.5 h sleep per night. Fatigue level peaks in the early afternoon and rises again sharply just before bedtime. In practice, however, fatigue is likely to increase as shifts get longer. The longer the shift, the less off-duty time is available for daily tasks (e.g., personal care, parenting, house- hold chores), and sleep may be sacrificed to accomplish these. Nighttime Construction Nighttime construction schedules of all variations show fatigue levels substantially higher than day schedules due to reduced sleep opportunity based on circadian pressure for wakefulness during the day (see Figure 6.9 for an example). Fatigue rises continuously throughout the night-shift work period and the commute home, peaking at bedtime (about 1 h after arriving home, according to the team’s models). Fur- thermore, nighttime construction schedules of all variations show a cumulative fatigue effect since reduced sleep hampers recovery (as in Figure 6.7 and, especially, Figure 6.8). Fatigue in night schedules is exacerbated by later work stop times and can be reduced through earlier stop times, such as 4:30 a.m. (Figure 6.14). Extended night shifts (10 h or more) tend to end later than shorter shifts and can result in severe sleep restriction (5 h or less) due to circadian pressure to wake around 1:00 p.m. For this reason, extended shifts should not be used on a regular basis for the same crew. Figure 6.13. Peak fatigue: day shift with normal sleep (7.5 h per night) and 1 and 2 nights restricted sleep (4.5 h). Monday Tuesday Wednesday Thursday Friday G re at er F at ig ue normal sleep 1 night, 4.5 hours 2 nights, 4.5 hours Monday Tuesday Wednesday Thursday Friday G re at er F at ig ue shift end 4:30 am shift end 6:30 am shift end 7:30 am Figure 6.14. Peak fatigue: night shift with various work stop times.

59 Taking naps is effective in reducing fatigue while working night shifts. A mid-shift nap on night schedules, even when short (30 min), is the single most effective fatigue counter- measure. It reduces peak fatigue and lowers the cumulative effect across days. A longer defensive nap (2 h) in the after- noon before the first night shift in a week is also helpful in reducing fatigue. Night shifts of any duration are substan- tially more fatiguing without these naps; a 5 × 10 night shift schedule is used as an example (Figure 6.15). In summary, a night shift schedule organized to accom- modate maximum recovery opportunity would end early and allow workers to take naps. Figure 6.16 compares peak fatigue for a typical day shift with “best-case” and “worst-case” night-shift scenarios, the best-case scenario being a shift that ends at 4:30 a.m. with workers taking mid-shift naps and a defensive nap, and the worst-case scenario being a night shift that ends at 7:30 a.m. and workers taking no naps. The best- case night shift scenario still results in peak fatigue that is at least double that of a typical day shift. However, by the end of the work week, the worst-case night shift scenario results in peak fatigue approaching twice that of the best-case night- shift scenario, as well as a substantially more rapid accumula- tion of fatigue throughout the week. Finally, the team’s fatigue models showed no substantive difference in fatigue levels for night-shift schedules where the worker reverts to a day schedule on days off (sleeping at night Monday Tuesday Wednesday Thursday Friday G re at er F at ig ue mid-shift & preventive naps preventive nap only no naps Figure 6.15. Peak fatigue: 5  10 (50-h week) night shift with and without mid-shift and defensive naps. Monday Tuesday Wednesday Thursday Friday G re at er F at ig ue day shift night shift, end 4:30 am, with naps night shift, end 7:30 am, no naps Figure 6.16. Peak fatigue: day shift versus best-case and worst-case night-shift schedules.

60 following a long morning nap after the last night shift) compared to maintaining a night schedule (sleeping 8 h dur- ing the day). However, it may be advantageous for workers to revert to a day schedule on days off, even if they will be return- ing to a night-shift schedule the following week, for two reasons. First, keeping a more “normal” schedule on days off will allow them to participate in many activities that are difficult while on a night shift, including family and social activities. Second, sleep quality for most individuals is poor during the day, even when the number of hours in bed would seem sufficient for adequate recovery. Little, if any, adjustment of the circadian rhythm to a night-shift schedule is expected unless such a schedule is main- tained for many weeks and light/dark schedules can be reversed. Other than indoor on oil platforms and in space, this is usually not feasible (Van Dongen, Belenky and Vila 2011). Shift Switching and Weekend Closures Granting workers a day off (i.e., a full 24 h between the end of one shift and the start of the next) when switching from a night-shift schedule to a day-shift schedule (or vice versa) is preferable to using double shifts or shifts with a very short break between. This is true for both mid-week shift switches and for short-term shift switching that occurs as a result of a weekend closure. For example, when workers who are usually on day shift are chosen to cover night shifts for a continuous weekend closure, a full 24-h break at each switch provides the best recovery opportunity. Managers and Designers Managers wishing to maintain high levels of on-site presence during weekend closures can reduce fatigue by engaging in two separate sleep periods (“anchor sleep” or “split sleep”)—a longer one of at least 4 h at night (the anchor sleep period), and a shorter one of 2.5 to 3 h (a supplemental nap) during the day (Mollicone et al. 2008). A manager the team inter- viewed reported his work and sleep periods during a recent weekend closure, and using this as a model, the team con- structed an anchor sleep schedule that would have allowed the same number of hours at work with regular presence on site during both day and night shifts. Peak fatigue could be reduced considerably using the alternative, anchor sleep schedule (Figure 6.17). Designers (or engineers) working high production sched- ules of 80+ h per week are vulnerable to cumulative sleep reduction and increasing fatigue. Fatigue levels are higher than for a standard day shift due to substantially reduced sleep opportunity while working very long (up to 14-h) days, and peak fatigue increases gradually throughout the week (Figure 6.18). Tactical countermeasures such as strategic naps and self-selected breaks can reduce the immediate impacts, but this type of schedule should not be sustained. Restricted Sleep A single night of sleep restriction leads to increased fatigue on the day shift for several subsequent days, and 2 nights leads to even greater fatigue (Figure 6.13). Acute sleep restriction (sleep loss) can occur for many reasons, including illness, household pressures, or emotional stress. Recovery frequently takes more than a single full night of sleep. Daytime construction schedules with unusually early start times (e.g., 6:00 a.m.) or long shift durations (e.g., 12 h) may result in curtailed sleep periods. Increased fatigue can be avoided by taking naps at mid-shift or after work. After-work naps should begin before 6:00 p.m. to avoid the circadian high-alert period that begins in the early evening. If naps are Friday Saturday Sunday Monday G re at er F at ig ue anchor sleep actual sleep Figure 6.17. Peak fatigue: manager’s actual versus possible anchor sleep schedule for 55-h weekend closure.

61 not used to alleviate excess fatigue, the peak fatigue trajectory of such a worker will be similar to that of high-production designers (Figure 6.18). Caffeine Use Caffeine can be used effectively before and during a shift for relief from acute fatigue. Caffeine is most effective when used sparingly; on day shift, it is most useful in the morning and during the “post-lunch dip” in early afternoon. Consumption should cease at least 5 h before bedtime, though there are large individual differences in caffeine impact. Sleep inertia experienced upon waking from mid-shift naps may be counteracted with consumption of a caffeinated beverage before the nap, which will take effect as the nap is end- ing (Reyner and Horne 1997; Van Dongen et al. 2001). This may be particularly useful on night shift, when workers may be concerned about their ability to awaken fully from a mid-shift nap. A cold, rather than hot, caffeinated beverage may facilitate rapid consumption prior to the nap. Table 6.2 provides a structured comparison of the sched- ules and countermeasure variations modeled, the major fatigue findings, and work practice implications. Summary of Work Practice Recommendations The most practical approach to work practice guidance for contracting firms is to use the findings from work schedule fatigue modeling to plan construction activities that incorpo- rate knowledge of fatigue’s impact on workers. Work practice guidance to address fatigue is a blend of specific tactics and countermeasure implementation, such as caffeine usage and worksite napping, and broader organizational practices associated with systematic evaluation and management of the problem. Unlike specific worksite problems such as traffic management or visibility, which can be addressed with proce- dures or technology such as lighting, work practices for fatigue management involve individual, crew, and organizational-level interventions. This intersection is best illustrated by the fatigue impacts associated with extended night shifts of 10 h or more. These shifts lead to severely curtailed sleep, because of reduced sleep opportunity and circadian pressure for wakefulness during the day. In general, the team recommends limited use of this type of schedule, since it leads to chronic sleep restriction during a multi-day schedule. Sleep restricted to 3.5 to 6.5 h per 24-h period, as modeled, leads to high levels of fatigue and cumulative effects on the worker. Thus, while construc- tion exigencies may require 12-h work periods or longer dur- ing a night closure, an organizational commitment to fatigue management would suggest that such a schedule be used only when it can be followed by an appropriate recovery period for the workers affected, or when other countermeasures can be implemented. More generally, the team would expect that the work schedule and work practice guidance described in this chap- ter and detailed in the Guide to Identifying and Reducing Workforce Fatigue in Rapid Renewal Projects can serve as a resource for project planners and work crew superintendents to manage the assignment of crew to specific shifts and con- struction tasks. By using the schedule guidance in combina- tion with awareness training and fatigue countermeasures, planners and superintendents can address fatigue issues such as cumulative sleep loss effects on the night shift before they become excessive. While the team does not suggest that con- struction projects be planned exclusively around worker fatigue management, the availability and use of fatigue Figure 6.18. Peak fatigue: high production designer or engineer schedule versus typical day schedule. Monday Tuesday Wednesday Thursday Friday G re at er F at ig ue standard day shift designer, high production

62 Table 6.2. Structured Comparison of the Schedules and Scenarios Modeled, the Major Fatigue Findings, and Work Practice Implications Scenario Major Fatigue Findings Work Practice and Countermeasure Approaches Day Shifts • No substantive fatigue differences across week or shift types • Fatigue increases to mid-afternoon, declines toward evening, increases before bedtime • Caffeine during day, but no later than 4:00 p.m. • Maintain consistent sleep and wake times throughout the week if possible • Maintain similar or identical sleep and wake times on weekend or non-work days • Strategic naps (on-the-job) to reduce impact of restricted sleep • Consume caffeine just before strategic naps to counteract sleep inertia on waking • Self-selected rest breaks to reduce fatiguing impacts of monotonous tasks or highly complex tasks Night Shifts: 5, 7, 8, 11-h closures • Sleep durations significantly shorter than day shifts because of circadian rhythm influences: 3.5 to 6.5 h due to circadian pressure to wake around 1:00 p.m. • Mid-shift nap substantially reduces peak fatigue within and across shifts, and reduces cumulative effects • Minimize use of extended shifts (10 to 12 h) due to reduced individual crew recovery opportunities • Caffeine during shift, but no later than 5 h before bedtime • Consider returning to day schedule (sleeping at least 8 h/night) on days off, following a morning nap on first day off from nights • Sleep in on the weekend to make up for sleep loss during the week • Strategic naps (on-the-job) to reduce impact of shortened sleep periods • Consume caffeine just before strategic naps to counteract sleep inertia on waking • Defensive nap in the afternoon before beginning night shift • Self-selected rest breaks to reduce fatiguing impacts of monotonous tasks or highly complex tasks • Supervisory monitoring for signs of fatigue and application of countermeasures Weekend Closure: 55 h • Modeling shows same effects as day and night shifts above • Field data suggest increased fatigue among day shift personnel in week following closure • Managers may feel they need to maintain a presence on the job site for as much as possible of the closure weekend; fatigue can accumulate during night shifts • Consider selective half or full day off after closure to provide recovery opportunity • Anchor (“split”) sleep schedule (nighttime anchor sleep and daytime nap) for managers to obtain 6 to 8 h in two separate sleep periods • Avoid double shifts • Use countermeasures appropriate for shift worked, as described above Switching Shifts • Modeling shows same effects as day and night shifts above • Avoid double shifts • Use countermeasures appropriate for shift worked, as described above Manager and Designer • Designers working high production can exceed 80+ h per week • Reduce high production designer workload through increased staffing and project planning • Same countermeasures as for day shifts, above Restricted Sleep • Schedules regularly leading to 6.5 h sleep or less nightly will result in cumulative fatigue • Sleep restricted to 4.5 h or less per night on one or two nights will result in increased fatigue levels, and this short-term sleep loss can affect fatigue long-term • In either case, fatigue level is higher than fatigue levels for standard extended day-shift schedules • For persons with consistently shortened sleep periods, a daily nap timed to avoid circadian high points (mid-shift or immediately after work) each work day will help maintain fatigue at low levels, and supplement the main sleep period • For individuals with acute fatigue from short-term sleep loss, sleep in on the weekend or take naps when able to make up for sleep loss during the week • Same countermeasures as for day shifts, above

63 management tools in conjunction with existing safety best practices should enhance safe and efficient project delivery. Fatigue risk Management Guide The materials assembled for this project can best be employed as a “toolbox” for end users such as construction project superintendents, state DOT inspectors, and project planners. To that end, the team has created a separate document as a companion to this report: the SHRP 2 Project R03 Guide to Identifying and Reducing Workforce Fatigue in Rapid Renewal Projects. The Guide is for safety managers, persons involved in cre- ating employee work schedules, DOT personnel, and others interested in applying specific and practical recommenda- tions for the management of worker fatigue on highway con- struction sites. The Guide contains four sections. Chapter 1 presents background and summarizes the main strategies for fatigue management. Chapter 2 describes the basic risk factors of rapid renewal construction schedules and the organizational processes and steps for implementing fatigue risk manage- ment. Chapter 3 describes the underlying physiology of human sleep and circadian rhythms, the fundamental mech- anisms that contribute to fatigue and work schedule interac- tions. Chapter 3 also provides a compilation of fatigue countermeasures (discussed in Chapter 4 of this report), incorporating those that are more effective, those that are less effective, and some that are in the preliminary research phase and not ready for widespread implementation. Chap- ter 4 contains specific shift schedule and work practice guid- ance for use by managers planning and executing projects. The team has also created two slide presentations, one for general highway workers and one targeted at managers, which can be used to train workers about the dangers and mitigation of fatigue in highway construction projects. These presentations are available at www.trb.org/Main/Blurbs/ 168766.aspx. Taken together, these four sections and the slide presenta- tions provide a resource for safety and training managers seeking more detailed technical information concerning worker fatigue and tools for implementing and evaluating components of fatigue risk management.