Abstract
BACKGROUND:
Airplane de-icing technicians work from either an open-basket or closed-basket.
OBJECTIVE:
The objective of this study is to identify the tasks that have an influence on the physical fatigue of open-basket aircraft de-icing technicians.
METHODS:
In a Canadian airport during the winter of 2016–2017, a field study was conducted in which the heart rate of 12 volunteer participants was collected. The data was analyzed along with the 22 tasks that make up the activity of open-basket aircraft de-icing. For each participant, the mean absolute cardiac cost per task was compared. The evolution of the cardiac signal based on the resting heart rate and steady state limit was also characterized.
RESULTS:
According to the cumulative results fatigue occurs for periodic tasks as well as double tasks. More precisely, the most physically fatiguing tasks are spraying de-icing and anti-icing fluids, moving the basket and truck, as well as tactile control and de-icing quality control at ground level.
CONCLUSIONS:
Similar studies would need to be conducted in other aircraft de-icing facilities to improve the generalization of the results.
Introduction
On a global scale, civilian air traffic is on the rise [1]. In North America, this trend was markedly confirmed between 2015 and 2016 according to [2] for the international market (+ 3.5% in number of passengers) and the domestic market (+ 4.9% in number of passengers). The northern hemisphere is exposed to heavy precipitations and temperatures below 0°C, which increases, for instance, the chances of ice forming on the external surfaces of aircraft [3]. The presence of ice on an airplane modifies its aerodynamics and the consequence of this could be fatal at the time of take-off [4]. Given this, Canadian legislation [5] makes it mandatory that critical surfaces of aircraft be de-iced prior to take-off and that staff be trained in the processes of ground and in-flight aircraft de-icing.
De-icing technicians perform ground de-icing act-ivities to make sure all ice has been completely removed prior to take-off [6]. The de-icing technicians drive de-icing trucks equipped with an open or closed-basket (also called elevating mobile platform) around the planes and de-icing and anti-icing glycol solvents. The health and safety of de-icing technicians must also be taken into account. This was made particularly evident in the accident of Royal Air Maroc’s Boeing 747 at Mirabel airport in Montreal, Quebec, Canada in 1995. The accident caused the death of three de-icing technicians and injured two others [7]. As mentioned by Nadeau et al. [3], it is a high-risk sector supported by high reliability organizations with few accidents.
Torres et al. [8] observed ground de-icing activities of 20 de-icing technicians and identified thirteen stress factors exerting an influence on work produc-tivity (e.g.: technological, economical, organizati-onal, ergonomic and safety factors). Ground de-icing technicians are exposed to several of these factors simultaneously, which leads to combined stress. Mor-eover, during a qualitative study using the NASA-TLX scale, Torres et al. [8] demonstrated that de-icing technicians perceive their work as being more physically fatiguing when performed in an open rather than closed basket. During a field study in the winter of 2016–2017, Landau et al. [9] observed 12 open-basket de-icing technicians. In their work, Landau et al. [9] describe the requirements imposed on de-icing technicians and the multiple interactions occurring in the work system of a centralized de-icing facility. In addition, during this field study, they identified 22 tasks performed during the open-basket de-icing activities. By measuring the level of energy turnover for each task, they established that the overall energy turnover of the workers varied between 4 and 13 kJ/min. For one of the two female participants studied, the energy turnover bordered the recognized upper limit (between 12 and 13 kJ/min for women).
According to the observations of the above-men-tioned qualitative and energy turnover studies, the work station of open-basket de-icing technicians co-uld be improved. To understand the origin of the perceived fatigue, the objective of the present study is to identify the most physically fatiguing tasks that make up open-basket de-icing activities to improve this occupational activity. Proper systematic risk ass-essment has been demonstrated as a fundamental and critical organizational element needed for proper work improvement investments [10].
Our results show that the most physically fatiguing tasks in open-basket airplane de-icing are: spraying anti-icing and de-icing fluids, special ground control, tactile control and moving the basket and truck.
Methodology
This research protocol involving human participants was approved by the research ethics committee with human participants of École de technologie supérieure in March 2017.
Population and variables studied
The population studied is composed exclusively of open-basket airplane de-icing technicians. The work of Landau et al. [9] is based on a field study conducted at a Canadian airport during the winter of 2016–2017. They describe the type of population studied and the conditions of observation of the 12 volunteer participants and open-basket de-icing technicians. The personal characteristics of the participants are compiled in Table 1.
Mean, maximum and minimum values of the personal characteristics of the participants
Mean, maximum and minimum values of the personal characteristics of the participants
The population of open-basket de-icing technicians in the facility studied is approximately 50 individuals in full season (between the months of December and March).
To identify the tasks that are physically fatiguing for open-basket de-icing technicians, the two variables used in this study are the list of tasks performed during open-basket de-icing and the heart rate of the participants.
With regards to data collection for the tasks performed during open-basket de-icing, the field study conducted by Landau et al. [9] provided 15 h 52 m of video recordings of the 12 participants, which correspond to 1039 registered activities, during the most intensive work periods of open-basket airplane de-icing. The work of Landau et al. [9] resulted in the break down and identification of the open-basket de-icing activity into 22 distinct tasks, as presented in Table 2.
List and numerical code of the 22, observable open-basket tasks presented in the research of Landau et al. [9]
List and numerical code of the 22, observable open-basket tasks presented in the research of Landau et al. [9]
The “unidentified tasks” are those that could not be identified using the video recordings, either due to poor visibility or a poor viewing angle. The “supple-mentary tasks” are tasks that are performed individually; they are either non-representative of the actual de-icing activity, or occur too rarely to base general conclusions for the entire group of participants. The “special ground control” task is performed outside of the basket, thus beyond the camera’s scope and is intended to assess the quality of de-icing at ground level. Finally, not all the tasks could be clearly coded, such as “radio contact” or “visual control”. De-icing technicians cover their face to protect it from harsh weather conditions. This makes the identification of tasks all the more complex. Using the recordings and the identification of these tasks, the de-icing activity was divided chronologically, at approximately one-second intervals.
To collect the heart rate data, during the field study with the 12 volunteer participants, all open-basket de-icing technicians’ physiological data was captured during the observation period using Hexoskin vests. These vests were adjusted to the size of each participant and worn directly in contact with the skin. The sensors of this vest made it possible to monitor the heart rate over time. As initially observed by [11], and employed by a good number of other researchers since [12], the heart rate is a reliable variable in the assessment of fatigue. However, as specified by [13], this variable can be affected by other stress factors other than physical activity such as temperature, mental stress or even other individual characteristics of the participants. In keeping with the recommendations of [14], the first 30 minutes of heart rate measurements were discarded to account for the possible initial emotional adaptive reactions of the test persons. Hence, the present study focuses on the evolution of the mean heart rate per second expressed as beats per minute (bpm) over the course of the de-icing activity. Once all the data had been gathered, the tasks and heart rates were temporally synchronized second-by-second, for all 12 volunteers.
To analyze the cardiac signal, two methods of data processing were used. First, the methodology for cardiac signal analysis similar to that used by Schlick et al. [15] was adapted for this study. Given the substantial amount of data captured over the entire period of observation (15 h 52 m), the data used for this study is limited to the most physically fatiguing work periods. Therefore, for each participant, the approximate values of the resting heart rate (RHR) before setting the steady state limit (SSL) of the heart rate were established. Schlick et al. [15] mention that when the heart rate of the participant exceeds the steady state limit, the individual is entering a state of fatigue. Using Schlick et al.’s [15] results, each participant’s SSL was calculated according to the type of resting position, by adding the following heart rate to the RHR: + 35 bpm in a sitting position or + 30 bpm in a standing position. This made it possible to accurately identify the tasks performed when the heart rate was higher than the steady state heart rate limit.
Second, a cardiac signal analysis was used that is based on a method initially introduced by [16]. Malchaire’s work proposes a method for interpreting recordings of the working heart rate according to the cardiac strain, or the absolute cardiac cost (ACC) when it is greater than the resting heart rate, required per task, for the entire group of participants. This variable is the mean amplitude of the heart rate signal greater than the resting heart rate. This method makes it possible to identify the tasks having the heaviest work load. Hence, this calculation was performed at the onset of each task for each individual participant. Then, as proposed by [17], the ACC greater than the resting heart rate per task was compared to the limits of cardiac strain, the values of which are presented in Table 3.
Cardiac strain limits greater than the resting heart rate as proposed by [16]
Cardiac strain limits greater than the resting heart rate as proposed by [16]
Finally, the results of these two methods of statistical analysis were compared. Given that few studies specifically analyze the tasks performed in the de-icing activity, the results obtained from analyzing the heart rate were compared to the observations in the energy study of Landau et al. [9].
Integration of the cardiac signal limits
For each participant, a graph was generated showing the cardiac signal for the entire de-icing activity for each airplane studied. Figure 1 shows an example of a graph illustrating the progression of the heart rate signal of an open-basket technician (alphanumeric code O7) during the de-icing of a CRJ9 airplane. This particular de-icing job lasted 7 minutes and 50 seconds.
Heart rate graph of de-icing technician O7, de-icing CRJ9 airplane.
The values of the resting heart rate (RHR) were measured while the participants were seated or standing. Then, for each participant and based on the RHR, the steady state limit (SSL) was determined. The length of the measurements of these cardiac limits is sufficient to enable a margin of absolute error that is between±1 bpm and 3 bpm for all 12 participants. The values obtained are presented in detail in Table 4.
Resting heart rate (RHR) and steady state (SSL), as well as the absolute error calculated for each of the 12 participants
Each participant underwent continuous observation for periods ranging from 59 to 96 minutes, which amounted to a total of 15 h 52 m for all 12 participants. However, the focus of this study is the work activities performed while in the open-basket, which for each participant amounted to between 13 and 58 minutes, and in total for all 12 participants 7 h12 m16s (or exactly 25,936s).
Moreover, once the resting heart rate and steady state limits were established, the length of time during which the heart rate exceeded both limits respectively was measured. To do this, the length of time during which the heart rate was greater than the SSL and RHR, over the total length of the time measured for open-basket work time only, was calculated. Table 5 shows the percentage of mean length of time during which the heart rate of the 12 participants is greater than the resting heart rate (RHR) and steady state limit (SSL), as well as the standard deviation and margin of error of this mean.
Results in percentages of time during which the heart rate for all participants is greater than the RHR and SSL
Results in percentages of time during which the heart rate for all participants is greater than the RHR and SSL
The table above indicates that the heart rate of the 12 participants exceeds the resting heart rate during an average of 89.7% of the work time in the open-basket. The heart rate of the 12 participants exceeds the steady state limit (SSL) 10.3% of the time on ave-rage (2397 seconds±6.26%) of the total work time in open-basket.
The periods during which the heart rate of the 12 active de-icing technicians was greater than the SSL were analyzed, which translates to 2397 seconds. We identified the duration (in seconds) of the various tasks performed during these 2397 seconds and illustrated the results in Fig. 2.
Distribution of tasks for which the heart rate is greater than SSL, expressed in seconds (heart rate of 12 participants > SSL during n = 2397 seconds).
It should be noted that the “Waiting in basket” period is rarely a time of passive waiting. The participants will for instance adjust their clothing, or perform tasks hard to distinguish on the video, such as listen to their radio communication headset or conduct visual controls.
Also, by taking into account our margin of error as presented in Table 5, based on Fig. 2, one can observe that the activities for which the heart rate of all the participants is greater than the SSL most of the time, are the following: spraying type-1 during 527 seconds (8 min 47s; ±6.26%), moving the basket during 551 seconds (9 min 11s; ±6.26%) and driving the truck using the basket controls during 431 seconds (7 min 11s; ±6.26%). The other tasks occur during less than 264 seconds, or 4 min 24s; ±6.26% of the time. Indeed, the category “Other” includes a set of tasks lasting a total of 192 seconds (or 3 m 12s; ±6.26%). These are tasks that seldom occur and when they do, it is over very short periods of time.
To complement these results, the mean absolute cardiac cost per task measured using the RHR for the 12 participants during 7 h12 m 16s (or 25 936s) of data was calculated. The results are presented in Table 6.
Comparison of the mean absolute cardiac costs for open-basket de-icing tasks, based on the limit of cardiac strain according to [16] (on n = 25 936s of data analysed for the 12 participants)
Comparison of the mean absolute cardiac costs for open-basket de-icing tasks, based on the limit of cardiac strain according to [16] (on n = 25 936s of data analysed for the 12 participants)
It seems that the tasks having the highest mean absolute cardiac cost are those that are performed simultaneously with another task. In fact, the double task of “Spraying type-4 fluid + Lowering the basket” is the task for which the cardiac strain is the greatest for the entire population of de-icing technicians studied, with an absolute error of approximately±2 bpm.
Currently, open-basket de-icing technicians feel their physical fatigue after their work day is greater than that of closed-basket de-icing technicians [8]. To the best of our knowledge, no other work has published analyses of the cardiac signal of open-basket de-icing technicians in view of identifying the most physically fatiguing tasks. The present study made it possible to identify these tasks as: spraying anti-icing/de-icing fluids, moving the basket and truck, tactile control, ground control and tasks performed simultaneously with other tasks.
The time constraint and number of participants are major limitations to the collection of data in the field study. Indeed, de-icing is a seasonal activity and the work environment of open-basket de-icing technicians is affected by, among other things, highly fluctuating meteorological conditions. It is difficult to anticipate when de-icing will be needed and to target the most fatiguing work in open-baskets.
The work environment of open-basket de-icing technicians involves multiple stress factors [8]. This study focused on assessing the physical load on the heart rate. However, it must be noted that heart rate may be affected by other stress factors other than physical activity, such as temperature, mental stress, dehydration or even the individual characteristics of the participants [15]. Thus, in light of the significant number of stress factors, and their constant variations, and the limited number of participants, it could be possible that other variables (e.g. weather, time of day, weekday, aircraft type, etc.) have an effect on heart rate.
Furthermore, this study encountered technical limitations. Given that the video recordings show the tasks performed from within the basket, all the tasks performed outside the basket were difficult to identify and analyze. In fact, certain tasks were difficult to clearly see and identify due to the position of the cameras and the limited size of the basket.
Given these circumstances, it would be relevant to conduct new studies for open-basket airplane de-icing. These studies could analyze respiratory data or the cardiac variability of the 12 study participants of Landau et al. [9]. Still, more studies similar to this one should also be conducted at other airport de-icing facilities to compare our results and improve their generalization.
The first results obtained using the methodology similar to that of [15], made it possible to identify the various tasks performed when the heart rate exceeded the heart rate’s steady state limit. However, Schlick et al. [15] specify that these limits are only indicators and that they depend on multiple factors (e.g. temperature, mental stress, dehydration or individual characteristics). Furthermore, it is difficult to compare the duration of these activities with each other, given that their duration depends on their frequency. For example, “Spraying type-4 fluid” or even “Special ground control” tasks are performed less frequently and for shorter lengths of time than “Spraying type-1 fluid”, because they are performed in different meteorological and organizational conditions.
When comparing the mean absolute cardiac cost, per task with the absolute limit of cardiac strain as defined in the literature by [17], one can observe that the tasks for which the work load was qualified as “rather heavy” or “heavy” according to the literature, are primarily those described as double tasks. While these intense tasks occur only occasionally, they generate a high and prompt absolute cardiac cost. However, the absolute margin of error of these results being±2 bpm, the hierarchy of these tasks among themselves is not strict. Thus, it possible to identify the cumulative effect of fatiguing tasks using these results, when these tasks are performed consecutively.
Finally, our conclusions were compared with the results of the energy turnover study of Landau et al. [9], using the same participants. The latter indicate that the tasks of special ground control and entering/exiting the basket are the most energy expensive tasks in the adopted posture (10 kJ/min), but involve some light hand work (between 1,75 and 3,25 kJ/min). The tasks of spraying de-icing and anti-icing fluids are also energy expensive (4 kJ/min) and they require that de-icing technicians work with their entire body (21 kJ/min). This study confirms Landau et al.’s [9] results, as well as identifies other fatiguing tasks such as tactile control and moving the basket and truck. Differences across these studies could be attributed to the effect of the multiplicity of stress factors on the cardiac signal, and the cumulative effect of fatigue.
Conclusion
With regards to the initial research question of this study, the tasks causing the open-basket de-icing technicians the most physical fatigue are the following single or double tasks: Spraying de-icing/anti-icing fluids; Special ground control; Tactile control; Moving the basket and truck.
Identification of the fatiguing tasks may help de-icing companies and airports to plan their areas of improvement for the processes of open-basket de-icing, to help reduce the physical fatigue of their workers. In future studies, it would be relevant to study the evolution of the cardiac variability or other respiratory data. Also, more studies similar to this one should be conducted in other airports to improve the generalization of the results.
Conflict of interest
None to report.
Footnotes
Acknowledgments
The authors would like to thank École de technologie supérieure and the Natural Science and Engineering Research Council of Canada for their funding support as well as the partner de-icing company and its workers for their participation.