Sunday, December 9, 2012

Appreciating the Role of Data in Atc: Part III - Highlights of a study on Fatigue

O stands for objective
The aim of the study as stated in the last post was to show that fatigue can be quantified according to Bayesian Principles; that factors present at the Atc units contributed towards fatigue and that there were organizational economic implications in addition to the safety undertones.

R stands for research
Literature showed that fatigue is not impossible to measure but it is a difficult condition to measure because of overlapping social constraints. The good thing about having a sound academic foundation in addition to the Atc background is that we will know what to look for based on readings and Atc experience. I used the research to develop a survey that centered on variables that will indirectly and directly capture individual controllers' perceptions about fatigue. 

A stands for analysis; C stands for confirmatory tests
I compared the results 2 logistic regressions to determine the likelihood or the probability of fatigue occuring given certain factors. 
Out of 22 invitations to nations of the Caribbean region, 50 Atcos from 8 nations responded with 2 of the controllers from the local associations of Barbados and Martinique that offered to help and give support till the end of the study. 
According to the demographics, 80% of the respondents received training in Area Procedural and Radar operations.  The majority of responses came from TBPB (Barbados), TFFF (Martinique) and TTZP (Trinidad & Tobago). The median age was 43 with half of the respondents married with 2 dependents. 20% of the respondents were females.
Using Likert scales to capture individual perceptions about controller fatigue, here are some illustrations of their responses:

The results from the first regression showed that as sleep loss increases, the likelihood of fatigue increases. Increasing domestic responsibilities also had a positive impact upon the likelihood of fatigue. The results from the second regression showed that as workplace culture increases, the likelihood of fatigue decreases. 

L stands for linkages made via inferences
For the study, I let workplace culture represent the physical environment: workstations and rest facilities; management policies as well as coping strategies for fatigue. 
The survey was open for 6 months, yet I only received 50 responses. True, these are islands so some units will have a handful of controllers. Yet the region is supposedly binded by economic and regional ties. How well is this reflected in Atc operations given the irony that TTZP( Trinidad & Tobago) is responsible for the upper limits of the air space in the Caribbean? 
I decided to use the legend of the Trojan Horse to illustrate that economics in the Atc unit also include weighing the costs of organizational policies, decision-making and in the case of the Caribbean, the effects of regionalism. 
I will discuss the trojan horse effect in the following post.

These are the limitations of the study:
i/ The study is limited by the size of the sample. My N=50 may be acceptable for the Caribbean but not for Atcos globally. 
ii/ There was a lack of data. With the appropriate data we can estimate the cost of fatigue or show the relationship between fatigue and the cost of incidents. 
How many Atc units of the Caribbean can readily present such data? Ironically, how many Atc units of the Caribbean can readily present records of airline companies who must be billed for using their airspace? 
iii/ Since fatigue is a subjective condition, it can be quantified as a latent construct in SEM analysis.

E stands for Exit decision
Workplace culture mitigates fatigue. Fatigue will be a recurrent issue in Atc units until it becomes an essential part of Atc governance.

Sorry about the late delivery of this post. Next post on Wednesday/Thursday! Boa semana o tudo!

Reddit Digg Share on Tumblr Delicious StumbleUpon Follow Me on Pinterest Print Friendly and PDF

No comments :

Post a Comment