Comparison of different Parameters of Heart Rate Frequency of Tasks Performed during Aircraft Open-Basket Ground Deicing Activities

Method Problem

For almost a hundred years, analyses of the heart rate have been used to assess individual stress, fatigue and recovery. There are a number of review articles on this subject, which also deal with the validity, objectivity or inter-rater reliability of heart rate measurements and their interpretation in case studies [1, 2, 3, 4, 5, 6, 7]. Particularly in the case of small practical case studies in companies, numerous factors influence the results. The contributing factors - e.g. climatic influences - can only be controlled or eliminated in rare cases. This was the reason for us to investigate a data set from aircraft de-icing - which indeed contains a wealth of contributing factors - with regard to its suitability for the interpretation of derived assessment measures of heart rate. We concentrated on Working Heart Rate (WHR), Heart Rate Reserve (HRR) and Absolute Cardiac Cost (ACC).

Real Problem

Under icy, snowy or cold conditions, plane de-icing is mandatory to insure safe plane take-off [8, 9]. Ground de- icing activities are performed by de-icing technicians who make sure all ice or snow contaminants have been completely removed prior to take-off [10, 11]. Staff are trained, tested and retested each de-icing season on processes, icing/anti- icing glycols, planes specific needs and others [11, 12]. In centralized de-icing facilities, plane de-icing involves up to four open or closed-baskets trucks that move around each plane [10]. Multiple interactions occur, making this work system a complex one [10]. De-icing organizations are high reliable organizations [12] focused on delivering an excellent and safe service to an airport (fast de-icing, clean plane) as well as keeping their technicians healthy and safe. When accidents occur [8], it could be fatal for de-icing technicians [8]. Technicians are exposed to multiple risks factors simultaneously of which basket-plane collisions and injuries of various kinds, i.a. including possible contacts with aircraft propellers [13]. There is also a whole range of possible work-related illnesses: discomfort or diseases of the musculoskeletal system, danger of severe frostbite due to weather-related effects of cold, snow, rain and freezing rain, noise and vibration induced diseases and others. A research report [10] provides an overview of the current state of research and the results of ergonomic/human factors studies in de-icing. Of the 22 tasks performed in open-basket de-icing, the physically fatiguing [10, 14] ones were identified and analyzed by Le Floch, et al. [15]. Time and motion studies of aircraft de-icing work were reported in Landau, et al. [16]. Energy turnover for each tasks was measured [16] and varied – among some peaks up to 21 kJ/min - between 4 and 13 kJ/min. More details and further information on risks of musculoskeletal disorders are reported in Nadeau, et al.

[17]. From our previous studies, one can conclude that open- basket de-icing work can be improved [18]. More precisely, more attention should be paid to work design and work organisation when spraying anti-icing and de-icing fluids, performing ground control and moving the basket and truck [15].

Consequences for Heart Rate Measurements

When it comes to de-icing, one must notice that we are dealing with a large number of possible influencing factors. The most important are:

• Type, severity and difficulty of tasks,
• Weather conditions,
• Other conditions of the physico-chemical working environment,
• Work organisation, in particular shift allocation,
• Type and number of aircraft per time unit,
• Operators’ characteristics, skills and abilities,
• Hormonal disturbances or stimulus conducting processes.

The number of factors influencing the heart rate is therefore large. Sammito, et al. [7] list a total of 17 major factors influencing heart rate and heart rate variability. They range from chronic alcohol consumption to the human circadian rhythm. If heart rate is to be interpreted in terms of workload, great caution is therefore required. Not all influencing factors mentioned in Sammito, et al. [7] are relevant for our study. Table 1 shows the main influencing factors. In field work, controlling all variables is difficult. In some cases, the sample sizes were too small to be able to apply statistical testing methods sensibly. Severe weather conditions and organizational changes were drawbacks for our initial sampling plan.

On the basis of the measurement data we examined the following questions:

• Are there trendy results when considering all operator studies?
• Does the type of the operator activities have significant impact on Working Heart Rate (WHR), Heart Rate Reserve (HRR) and Absolute Cardiac Cost (ACC)?
• Does the sequence of activities during each study influence WHR, HRR and ACC?
• Are there (linear) trends in heart rate frequency indicating central fatigue?
• Given the large number of influencing variables (see above), are heart rate field studies at all suitable for evaluating a relatively?

Working Heart Rate and Heart Rate at Rest

To eliminate part of these different influences in field studies, it is advisable to deal with the heart rate increase or the working heart rate (WHR).The working heart rate is the difference between the current heart rate and the heart rate at rest:

WHR (t) = HR (t) – HRrest

Resting heart rate is usually determined prior to a strain study in a break or lab room. However, for organizational and safety reasons, it is not always possible to measure the resting heart rate prior to the start of the data collection. In those cases, one can use the minimum heart rate value of each participant during the study as a resting value.

Depending on the posture adopted, the continuous endurance limits are determined. The amount of workload to be handled without increasing fatigue (i.e., a steady-state condition) is referred to as a continuous endurance limit [28]. If the load exceeds this steady- state value, fatigue progresses. The maximum endurance limit is defined in ergonomics as the highest possible performance, which can be provided over a whole working day without additional rest periods.

When measuring heart rate at rest, the participant can lie, sit or stand. If the resting heart rate is measured while lying down, one assumes an energetic continuous endurance limit of 40 beats per minute [20]. If the resting heart rate frequency is measured while sitting, the continuous power limit is 35 beats per minute, while standing, 30 beat per minute. All values refer to an 8-hour shift. These limits are only to be considered as rough reference points, as the physical characteristics and training state of the person influence the continuous endurance limit.

A largely linear relation exists between heart rate and oxygen consumption [19, 29]. Thus, endurance limits correspond to work energy rates of approximately 17 kJ / min in males and 12 kJ / min in females.

Heart Rate Reserve

The heart rate reserve (HRR) is sometimes calculated instead of the WHR [22]. HRR is a validated objective measure of the balance between the work demands and resources of the worker (“Karvonen formula”) [5].

$$ H R r e s e r v e = \frac {H R w o r k - H R r e s t}{H R \max - H R r e s t}. 1 0 0 $$ The physical strain is then expressed with the current heart rate as a percentage of the HRR. As confirmed knowledge, the threshold of 33% HRR is used for the maximum energetic load during an 8-hour workday [30]. Working below this threshold will most likely lead to a healthy balance between work demands and physical resources. Meanwhile, exceeding the threshold will result in fatigue and exhaustion in many workers.

Absolute Cardiac Cost

Related to WHR and HRR is the calculation of the absolute cardiac cost (ACC): beats per min above resting heart rate (RHR):

ACCi = HRi-RHR

HRi is the heart rate averaged over a certain period of time. It makes sense to average over the respective operator activities i.

Frimat, et al. [6] has defined the following range limits of ACCi (Table 2):

ACC thus differs only slightly from WHR. While we related WHR to the instantaneous heart rates f(t), we calculate ACC for the individual operator activities. For WHR we determine the exceeding of energetic continuous endurance limits. For ACC, on the other hand, we perform a classification according to Frimat range limits.

Relative Cardiac Cost

Besides the absolute cardiac cost one can also compute relative values. It is useful to refer to the 99th percentile of the frequency distribution of the heart rate frequency [31]. This means that the peak stress of one percent of the heart rate frequency is ignored. Relative cardiac cost according to the Karvonen formula mentioned above is a percentage value with Meunier [31]:

$$ R C C _ {99 \%} = H R _ {99 \%} - \frac {H R _ {rest}}{H R _ {\max} - H R _ {re}} $$ [ ] 99% 99% max *100 % rest rest HRmax is calculated according to Astrand, et al. [26] to HRmax = 220-age±10% where “age” is the age of the subject. There are approaches to improve this simple formula, but they have not always been successful [32]:

HRmax = 207-0.7*age

The continuous endurance limit of RCC is stated by various authors as follows (Table 3)

Recovery Heart Rate

The recovery heart rate can also be used as an indicator of previous exercise. It characterizes the recovery ability of the cardiovascular and metabolic system. It decreases exponentially after the end of exertion. However, since the employees continued to work after our measurement intervals and no rest phase occurred, we did not evaluate the recovery heart rate in our study.

Heart Rate and Test Subjects

Working Heart Rate (WHR): The following box plot diagram shows the frequency distributions of working heart rate WHR (Figure 1).

Click to enlarge

Arithmetic mean is between 16.9 and 45.0 bpm for samples O1 to O11, for the whole data set 28,95±12,28 bpm. O1 and O4 have the highest mean; O5, O6 and O8 have the lowest mean of working heart rate.

We calculated the working heart rate frequency and tested for transgression of the threshold 35 bpm. Tab 8 shows the number of exceedances WHR >= 35 bpm [19, 20] for the individual probands. For O1, O4 and O7 it is likely that the 35 limit value is often exceeded during the shift. Of course, physical fatigue or even cardiovascular damage cannot yet be inferred from this. Short-term exceeding of the limit values can be compensated by a subsequent recovery phase.

On average, in about 30.4 % of cases, the threshold 35 bpm is exceeded. Noteworthy is the large fluctuation range (from 3.4 % up to 63.9 %) in the individual experiments.

Heart Rate Reserve (HRR): Arithmetic mean is between

0.206 and 0.573 for samples O1 to O11, for the whole data set 0.372 ±0.158. The following box plot diagram shows the frequency distributions of heart rate reserve HRR (Figure 2).

Click to enlarge

13,725 out of 25,711 observations (=55.99 %) show high aerobic load with respect to HRR (Table 9). Again, there is the large fluctuation range (from 8.2 % up to 95.8 %) in the individual samples.

Absolute Cardiac Cost (ACC): Figure 3 shows that mean values of the ACC are in the light to moderate range of the physical strain (according to the Frimat-classification) [6]. Of course, it should be borne in mind that the averaging process compensates for phases of high physical stress by means of periods of lower stress. From the point of view of organizational work design, this is desirable. Table 10 below, however, refers to ACC peak values that are occurring regularly in the workflow.

Click to enlarge

Table 10 show the percentages of measurements in studies O1 to O11 exceeding the respective limit values. The HRR and ACC long-term exposure limit values are exceeded in about 43 % of all cases. In contrast, in the case of the WHR, only 29.3 % exceedances (O1, O3, O4, O7) are present. HRR and ACC therefore lead to the more cautious limit values. It should be noted that O1 and O11 were exposed to exceptional meteorological conditions.

The large discrepancies between instruments at subjects O2, O5, and O6 are particularly noticeable. The HRR becomes so high because the denominator of HRR (HRmax–HRmin) is low. Possible reasons for this could be:

• Test subjects have only a limited maximum capacity of heart rate frequency,
• During the relevant studies only less demanding activities were carried out,
• The mode of operation of these subjects is very economical.

It is therefore clear that it makes sense to look at the average values of characteristic values of the working heart rate if one wants to check whether the cardiovascular system is in balance in the long term. If, on the other hand, one discusses the percentage values for exceeding threshold values, then this is a statement about stress/strain peaks during the shift.

Heart rate and activities: As it was shown above that HRR and ACC are particularly cautious measures of physical exertion, we limit our discussion of the activities to their analysis. Table 11 shows the statistical analysis of HRR and ACC for deicing activities.

The following activities lead to high HRR/ACC values (Table 12):

In most cases, these are activities that lead to a superposition of forced postures and dynamic use of the upper body. Informational and emotional strain also plays a role. Even during visual inspection (e.g. tactile control), the operator must bend forward strongly and steer the basket sideways at the same time.

In Table 11 there is only a partial overlap between high HRR and high ACC levels. There is a case for using the HRR measure rather than the ACC measure, when considered cautiously (protecting the employee).

Fatigue over the Shift Time

Linear trends were calculated for the time-dependent heartrate over the working shift. From a (possible) increase of the trend lines, the central fatigue of the operator can be assumed [1].

For the absolute value of the heart rate, the equations for a linear trend were determined (Table 13).

It turns out that with the exception of O1, O2 and O5, a positive increase in the heart rate can be assumed, i.e. that we can assume increases in central fatigue.

Operator Workload Derived from Shift Characteristics

Parameters of heart rate and limit criteria were discussed with respect to operator workload (OW) (Table 14).

We have tentatively defined:

. . * No of aircraft during observation period OW No operatorsin shifts observationtime = Table 14 shows operator workload OW for the individual test subjects.

If one checks for the correlations of high cardiovascular strain (ACC: rather heavy ...intense, HRR>33%, WHR>35 [min-1]) with OW, negative correlation coefficients are obtained (Table 15) (All three variables are normally distributed (Shapiro-Wilk-test, ρACC = 0.0509; ρHRR = 0.906; ρWHR = 0.423)).

These non-significant results indicate that the number of aircraft in the shift and the number of operators available cannot be used to draw conclusions on operator strain. Other variables, such as weather, process organisation, quality of the shift crew, etc., obviously play a major role.