Aviation Psychology

Pilot Selection

Pilot selection procedures, world over,  were started as a result of high level of attrition and pilot wastage. Initial tests were physiological and psychological. Later on psycho motor tests along with cognitive tests were  introduced.  Royal Air Force (RAF) developed Basic Attributes Test (BAT) , which was considered as the origin of aviation psychology. When RAF introduced the test, the wastage rate at the training academy came down from 48% to 25%.  The test included objective paper-pencil as well as mechanical aptitude tests. In the 80s, with the advent of computers, reliable testing for flying aptitude could be made possible. In 1984, computerized tests started. In modern aircraft sensor, control, communication and weapon delivery systems are highly automated with accuracy and precision with the help of on board computers. In the cockpit, the pilot not only takes up the job of flying,  he also functions as a manager of the system. Under critical situations (For example during deep penetration in enemy territory) he may have to take a decision, where computers are unable to decide or their decision must be overridden. It is the man's reprogramming ability that makes the neutralization of the new emerging dangerous factors. Man's superiority in decision-making is superior through his pattern recognition ability and his intuitive experiential knowledge. An expert in the cockpit adds to the multiplier effect of the system. The pilot gets this expertise from the training academies. The pilot has to receive multi channel inputs, read alphanumeric displays, choose pertinent data selectively, maneuver through the controls, make choices and decisions during dogfights and otherwise, apply intuition build by long standing experience to act in unpredictable situations. Increased automation does not decrease the mental workload, it in fact increases it. 
The United States Air Force (USAF), which was earlier using electromechanical devices for assessing the psycho-motor abilities have now developed computerized versions of these tests. The Belgium Air Force originally used a test battery consisting of nine psycho-motor tests in its pilot selection, which were reduced to three test battery after validation. Flight Aptitude Selection Test (FAST) and Revised Flight Aptitude Selection Test (RFAST) of US army aviators tested spatial ability, mechanical ability and aviation information. Royal Air Force (RAF) in England has the largest Computer Based Assessment (CBA) center for pilot selection since 1985. This contained (1)  control of velocity test (CVT) of anticipatory tracking (keeping a cursor on a moving track), (2) sensory motor apparatus (SMA) test of compensatory tracking (keeping a moving cursor centered on the screen), and (3) instrument comprehension B (INSB) test of speed of visualization. Pilot selection has become essential for the following factors.

1. To lower the accident rates
2. To curb the training costs
3. In the wake of modern aircraft, the role of pilot has shifted from perceptual/psychomotor ability to information procession ability.

Pilot Selection through Computerized tests:

1. General Aviation Trainer (GAT – I) by Hill and Goebel in 1971
2. Automated Pilot Measurement System (APAMS)  by by Long and Varney in 1975
3. Canadian Automated Pilot Selection System (CAPSS) in 1989
4. Heavy jet simulation testing for ab initio pilot candidates by Scandinavian Airlines System (SAS) in 1991
5. Basic Flight Instruction Tutoring System (BFITS) using PC by Tirre 1992
6. Canadian Automated Pilot Selection System (1999) using  a single engine light aircraft flight simulator testing accuracy in maintaining prescribed flight parameters, speed of response to errors or warnings, variability in performance, smoothness of operation and avoidance of over corrections, and coordination of flight controls.  
7. A combination of paper-pencil and computerized tests in Belgium testing reasoning, spatial representation, verbal factor, organization and memory test in 2000
8. A test of mechanical principles and devices, mathematics, acoustic memory, capacity and control of attention, perceptual speed, spatial comprehension, psycho motor and multiple task capabilities in Germany in 2000
9. Visser (2000) reports that pilot selection in Netherlands comprised verbal reasoning, non verbal reasoning, logical arithmetic, technical aptitude, flexibility, achievement orientation, knowledge in the desired field of employment. along with six flight simulation tests (GVSS)  
10. Aviation Selection Test Battery (ASTB) consists of four tests: the Academic Qualification Test (AQT), the Mechanical Comprehension Test (MCT), the Spatial Apperception Test (SAT), and the Biographical Inventory (BI) 
11. Air Force Officer Qualifying Tests (AFOQT). The AFOQT consisted of verbal analogies,  Arithmetic Reasoning, Reading Comprehension, Data Interpretation, Word Knowledge, Math Knowledge, Mechanical Comprehension, Electrical Maze, Scale Reading, Block Counting, Table Reading, aeronautical knowledge, Rotated Blocks, General Science and Hidden Memory.
12. Basic Attributes Test (BAT) of USA is a computerized test consisting of 15 tests -- two hands coordination, complex coordination, psycho-motor skills, item recognition checks, short term memory, time sharing, activities interest inventor. The test comprehensively tests cognitive, affective and psycho-motor abilities.
13. Alternate Flight Aptitude Selection Test (AFAST) consists of seven sub tests -- background, instrument comprehension, complex movements, helicopter knowledge, cyclic rotation, mechanical functions and self description.

Advantages of Computer Assisted Testing for Pilot Selection:

1. No need for data entry at any stage hence eliminates the errors in this process
2. On-line scoring with highest objectivity, accuracy, and reliability
3. Response latencies can be measured in the order of micro seconds on hard real time systems
4. High acceptance rate due to visual effects
5. New dimensions can be measured, which were impossible to measure using conventional tests.
6. Application of new computation intensive statistical models, testing theories is possible for further R&D.

Procedure for development of the test:  The general procedure which is required to develop any computerized testing system is

1. Job Analysis:  To identify skills, abilities, competencies, interests, behaviors and personality traits ie knowledge, skills, abilities, and other characteristics (KSAOs) required to perform the tasks so that a selection procedure and methodology can be formulated to determine how well the job applicants fit the requirements for job. Job Analysis is the process of collecting & analyzing information about jobs to write:  (1) Job Description: a document that identifies the tasks & duties performed by a job (2) Job Specification: a document that identifies the qualifications required by a job.  There are various methods and integrated way of using these will give comprehensive set of aptitudes / abilities required for the job. The methods are
1. Narrative : Collecting information about task, duties, qualifications for the job from incumbents and supervisors
2.  Time & Motion Study
3. Structured Job Analysis
1. Functional Job Analysis: 7 scales measuring tasks, education, functions
2. Position Analysis Questionnaire (PAQ)
3. Professional and managerial position Questionnaire(PMPQ)
4. O*NET : Occupational Information Network Database
5. Task Inventory Procedure
6. Critical Incidents Technique 
7. Ability Requirements Scales  
8. Personality-Related Job Analysis Procedures 
9. Cognitive Task Analysis
4. Future oriented strategic job analysis
5. Competency Modeling - finding out critical success factors which should be tied to organization strategy and objectives
2.  Finalization of aptitudes / abilities required for the job: The comprehensive abilities / aptitudes can be operationally defined and then can be evaluated by subject matter experts (experienced pilots and instructors) to rate them in terms of importance / relevance for the job.  This procedure may give the broad and specific aptitudes required for the job. Pilot tasks can be broadly classified into six categories namely 
1. Cognitive tasks 
1. Numerical ability
2. Verbal ability
3. Non-verbal ability
4. Spatial ability
5. Mental arithmetic
6. Mental Rotation
7. Job-specific cognitive ability like instrument comprehension,
2. Tracking tasks 
1. Compensatory (keeping a moving cursor centered on the screen)
2. Anticipatory (keeping a cursor on a moving track)
3. Pursuit (Following a moving target) 
4. Dual Tracking (Tracking two targets simultaneously)
3. Reaction
1. auditory stimulus
2. visual stimulus
4. Memory
1. Recall
2. Interference
5. Estimation
1. Size
2. Velocity
3. Landing judgement
6. Visual scanning
1. Vigilance
2. Situation Awareness (e.g. object estimation test) 
7. Time sharing ability
1. Various combination of above mentioned tasks can be combined together, especially tracking and reaction tests.
3. Road-map for development of the Tests:  
1. Development of cognitive test items for multiple choice tests using the classical test theory (CTT) or still better and more efficient the Item response theory (IRT).  This will standardize the items, items parameters and the administration procedure. These tests can be computer administered (in CTT) or computer adaptive (in case of IRT). The developed tests can be administered on an appropriate sample and standardized. The reliability, validity, information content, SEM and lack of bias can be ascertained.
2. Development of tracking tests, reaction time tests, memory tests, estimation tests, visual scanning tests, memory tests can be developed to capture the specific ability. To establish the correctness of the scores, a hard real time system is the absolute necessity. Scoring points for specific events (eg superposition of target, various kind of errors and omissions) can be decided based on regression weights or by fitting a proper distribution on the scores as well as on the specific events.
3. Analysis: Composite battery is administered on representative samples and concurrent validity has to be found. Attenuation factor due to low reliability of the criterion measurement test (in order to take care of its fallibility) needs to be taken into consideration. This is achieved by dividing the validity coefficient by the square root of reliability coefficient of criterion measurement test. Regression analysis can be carried out to find significant predictors and discriminant analysis can be carried out to find an initio clusters of the subjects. 
4. Norms development and Time Fixation: Norms provide a basis for comparing the individual with a group. Common formats for these norms include percentile ranks, z scores (calculated as (x-mean)/SD), sten scores (calculated as (x-mean)/SD*2+5.5, so that these have mean 5.5 and SD of 2), and other forms of standardized scores. Time Fixation can also be done by aligning the score distribution with the actual test time distribution and then further aligning the desired locations on these two distributions.If group differences among the test-takers are found between the scores then different norms and administration time can be fixed for the different groups. The scores can also be factor analyzed and the composite scores for various factors also can be used for further analysis or validation or for short listing.

   Z-scores	< -2.0	-2 to -1.5	-1.5 to -1.0	-1.0 to -.5	-.5 to 0	0 to +.5	+.5 to +1.0	+1.0 to +1.5	+1.5 to +2.0	> +2.0
   Percent	2.3%	4.4%	9.2%	15.0%	19.2%	19.2%	15.0%	9.2%	4.4%	2.3%
   Sten score	1	2	3	4	5	6	7	8	9	10

5. Criterion Validation: The tests can be administered to trainees and the training marks can be regressed against the scores of the tests to find out the significant predictors. Discriminant analysis can then be carried out to verify the clusters. Regression analysis may give many different sets of predictors of comparable predictive power. Appropriate care must be taken in this case to isolate the particular set after doing the causal analysis. Other classification methods like ANN(artificial Neural Networks coupled with faster training algorithms like simulated annealing or genetic algorithms) may be considered in this case to enhance the classification and the predictive power.
6. Finalization: The test battery can then be finalized by keeping only significant predictors and weeding out non significant ones. Norms development and time fixation can be reworked. 

Pilot Training

Pilot Mental Workload

Workload can be defined as the portion of human resources an operator expends when performing a specified task. Since resources are limited so overload can lead to performance decrements. Workload is assessed by subjective methods, secondary tasks, physiological techniques and measures of primary
task performance.  The techniques for measuring workload are characterized by five criteria namely- in order of importance (1) Sensitivity (capability to discriminate significant variations in workload imposed by task or group of tasks), (2)  Diagnosticity (the capability of a technique to discriminate the amount of workload imposed on different operator capacities or resources e.g. perceptual vs. central processing motor resources), (3) Intrusiveness ( the tendency of a technique to cause degradation in the ongoing Primary task performance, (4)  Implementation requirements ( the factors related to the ease of implementing a particular technique e.g., instrumentation requirements and operator training requirements etc and (5) Operator acceptance ( the willingness on the part of operators to follow
instructions and actually utilizing a particular technique). Workload is not only task specific but also person specific. Individual capabilities, fatigue, monotony, age, experience, motivation strategies to perform, mood, operator state, environment demands, aircraft ergonomics, automation, feedback etc will all affect the experienced workload.

1. Subjective methods / Rating Scales: These are classified according to the level of measurement: nominal, ordinal, interval and ratio scale level. Most rating scales have ordinal or interval scale level. On an ordinal scale level non-parametric statistics must be used, whereas on interval scale level the parametric methods can be applied. The scales can be uni-dimensional, multidimensional or hierarchical. To get diagnostic information one has to use multidimensional ratings of workload (e.g.NASA-TLX), which have several subscales and can detect the workload drivers. Hierarchical rating scales (e.g., Modified Cooper-Harper Scale) have the advantage that the rating is done in a step by step procedure with a decision tree, which makes the evaluation process easier. Their disadvantage is that they give no diagnostic information. The advantages common to most workload rating scales are High validity,  High face validity, Minimal costs, Easy to implement. Non-intrusive, Can be used prognostically. The disadvantages are Rating scale results can be influenced by characteristics of respondents, like biases, response sets, errors, pretest attitudes, faking etc

1. NASA Task Load Index (NASA-TLX): Workload emerges from the interaction between the requirements of a task, the circumstances under which it is performed and the skills, behaviours, and perceptions of the operator. Workload is regarded as a multidimensional construct so that consequently a workload measurement method must be made up of multiple workload dimensions. It has an individual weighting procedure for the different workload dimensions, which is assumed to reduce between subject variability of the results. The six subscales can be divided into three groups. Characteristics of the task: Mental, Physical and Time Demands. Behavioural characteristics: Performance and Effort. Individual characteristic: Frustration. Each of the bipolar subscales of NASA-TLX consists of 20 five point steps from 0-100, the endpoints having verbal descriptors. There are two phases of scoring. In the first phase the subject rates the task with regard to workload. In the second phase each subject makes a paired comparison, deciding with all 15 possible pair combinations of the 6 dimensions which pair element is more important with regard to workload in the rated task. From the results a rank order of the dimensions from 0-5 is derived by which the individual subscale scores of the rated task are weighted. By summing up the weighted subscale scores and dividing them by the sum of the weights (= 15), the Mean Weighted Workload Score is obtained, that indicates workload in per cent. The scale has good face validity, interval scale assumption, acceptable diagnosticity due to six scales. 
2. Subjective Workload Assessment Technique (SWAT):  Subjects rate the workload of a task on the basis of the dimensions of "time load", "mental effort load", and "psychological stress load".  The method uses conjoint measurement and scaling and the ratings on these
   dimensions, which are made on an ordinal scale level can be combined into one overall workload score on an interval scale level. Each of the three subscales has three levels in the form of a category scale with verbal descriptors that outline each dimension. When applying SWAT one can distinguish two phases: scale development phase and event-scoring phase. In scale development the subject has to rank order all 27 combinations of the three levels of each of the three dimensions according to her individual perception of increasing workload. For this purpose the subject uses 27 cards with descriptions of these 27 combinations. After that, it is necessary to test with a computer program if the card sort is
   in accordance to a series of mathematical axioms so that conjoint measurement and scaling can be applied. After this procedure the SWAT technique uses conjoint analysis to convert the rank data of the card sort into an interval scale from 0-100 in which each of these 27 combinations has a fixed value. After scale development event scoring can take place, in which the subject evaluates the relevant task with regard to time load, mental effort load, and psychological stress load (e.g., 2, 1, 3). The scale value associated with this combination (e.g., 30.5) is the dependent variable used in the further analysis. The advantages are interval scale measurement, time-economic, prognostic value etc. 
3. Modified Cooper-Harper (MCH) Scale: The Cooper-Harper Scale has been used successfully for evaluation of aircraft handling tasks and other motor tasks. In more advanced complex automatic systems the human operator's role has changed and he is less involved in active control of the system and more occupied with activities like perception,
   monitoring, evaluation, communication, and problem solving. As the original
   Cooper-Harper Scale was not developed for evaluation of such activities, Wierwille and Casali modified this method, so that it can be utilized with tasks, characteristic for modern systems. The derived Modified Cooper- Harper (MCH) Scale retains the decision tree of the original scale but has changed the wording.  The MCH Scale has been validated in a simulated flight environment with perceptual, mediational (cognitive) and communication tasks. The authors assume that the scale will also be sensitive in other environments with
   such tasks, typical for modern operator machine systems. The scale has a low diagnosticity. and the measurement is on ordinal scale level.
2. Secondary Task Methods: The method is based on the assumption that the human operator as an information processing system has only limited channel capacity and the spare capacity left by the primary task is filled up by the secondary task. The higher the performance decrement in the secondary task the lower is the remaining spare capacity of the operator in the particular primary task and the higher is operator workload. In contrast to the external secondary tasks, the embedded secondary tasks (e.g., radio communication when flying an aircraft) are part of normal system
   functions. They appear as a natural and integral part of the task of interest and reduce the main problems of conventional external secondary tasks: task intrusion, poor operator acceptance, and high implementation demand.Workload measurement is secondary task specific: which aspect of workload is measured depends on the selected secondary task. When using different kinds of secondary tasks with the same primary task, the results can dissociate. Such secondary tasks should be selected, which require the same resources necessary for performance of the primary tasks. Some tasks produce such a high workload (e.g., mental mathematics) that a secondary task cannot be added. Secondary tasks require a sufficient amount of training without and with the simultaneously performed primary task. The possibility of secondary task intrusion on primary task should always be considered carefully. Secondary tasks measure on interval scale level. Use of parametric statistics is possible. Th advantages of secondary tasks are Objective assessment with minimal influence of subjective factors, Good diagnosticity when using different secondary tasks, Measurement of continuous workload possible, interval scale measurement. The disadvantages are (1) Secondary tasks measures are task specific. Results from different main tasks but with the same secondary task cannot be compared, as interaction between main tasks and secondary tasks is different (2 Studies with secondary tasks show contradicting results. For most tasks there is one study showing an improvement, another degradation, and a third no change of the
   secondary task performance with increasing workload.
3. Physiological Workload Measurement Techniques: These are based on the assumption that workload is reflected in the physiological responses of the central nervous system, the autonomic, the somatic and the endocrine system.  What has been said in respect to secondary tasks is also valid for physiological measures of workload: every technique has been shown to be sensitive to
   workload and almost every technique has been shown to have failures. From this the conclusion should be drawn that physiological methods should only be used in addition to subjective workload
   measurement methods. Different physiological methods (and also measures) measure different aspects of workload. Physical activity can influence physiological measures (e.g., heart rate). Th advantages are (1) Objective assessment of workload without involving subjective aspects is possible (2) Many different methods can be applied simultaneously (3) The subject cannot voluntarily influence the result (like with rating scales). The Data analysis can be demanding, being both in time and frequency domains. The mass are
1. Heart rate and heart rate variability
2. Measures of eye functions, such as eye movements and eye scanning, blink rate and blink duration, pupil diameter and pupil dilation provide measures directly related to visual workload.
3. Measures of brain activity: Spectral analysis of the electroencephalogram (EEG), Evoked
   Cortical Potentials (P 300 amplitude of the ECP) are relevant to perceptual task demands and are sensitive to aspects of cognitive processing
4. Measures of body fluid analyses: Salivary cortisol levels significantly differentiated between different levels of stress and workload.
5. blood pressure 
6. electromyography (EMG) 
7. galvanic skin response (GSR)
8. critical flicker fusion frequency (CFF)
4. Primary Task Measurement:  Primary task can be defined as maintenance of safe trajectory and control over the aircraft in simulated and real environment. The task is largely a psycho-motor task involving eye-hand coordination and involves tracking-ability. Different measures of performance can be taken e.g. number of errors made, reaction time, speed of performance, deviations from defined path (measured by SD-standard deviation from the position), standard deviations of various controls from mean position (RMS error can me a metric in this case), Time taken to neglect path errors, time to finish the trajectory. composite score etc. SDs of baseline and the mental loaded conditions can be compared to arrive at a threshold level. Deviations in SDs can be computed to assess mental workload. The workload found by various methods can be combined to give a composite index of effects. The disadvantages of this method is that Performance on the primary task may be insensitive to workload change if operators compensate by increased effort.

Air Safety / Accidents of Aircraft

Human error is the primary contributor to more than 70 percent of all aircraft accidents (both commercial and military). Pilot error is a far more likely cause of an airplane crash than mechanical failure or bad weather conditions. The vast majority of aviation accidents occur at a relatively low height – either before, during or after take off or landing and half of all air crashes occurred during approach and landing. There were quite a number of accidents where a perfectly normal airplane suddenly crashed into a mountain without the pilot realizing he was about to crash. Pilot errors were often cited here as the cause of such crashes.

Swiss Cheese Model

In his "Swiss cheese" model of human error, Reason describes four levels of human failure, each influencing the next. The first level depicts those Unsafe Acts of Operators that ultimately led to the accident. More commonly referred to in aviation as aircrew/pilot error, this level is where most accident investigations have focused their efforts.  Represented as "holes" in the cheese, these active failures are

typically the last unsafe acts committed by aircrew. For example failure to scan the instruments correctly. There may be other latent factors in the causal sequence leading to accidents, which may lie dormant or undetected for days, until one day they adversely affect the unsuspecting aircrew. Within this concept of latent failures, Reason described three more levels of human failure. The first involves the condition of the aircrew as it affects performance. This is referred to as Preconditions for Unsafe Acts. This level involves conditions such as mental fatigue and poor communication and coordination practices, often referred to as crew resource management (CRM). In many instances, the breakdown in good CRM practices can be traced back to instances of Unsafe Supervision, the third level of human failure. For example, two inexperienced pilots for night flying without proper CRM training coupled with aircrew errors will magnify the problem. Organizational influences, at the last level, can impact the performance too. For example, fiscal austerity may force the supervisors to task "non-proficient" aviators with complex tasks. Swiss Cheese model does not define what the "holes in the cheese" really are. These may be defined in the context of the system under investigation. Hole in one layer, may be covered by good practices in the next layer, but once the holes in all the layers are aligned, accidents happen.

The “Swiss cheese” model of human
error causation (adapted from Reason, 1990).

 HFACS Model

Drawing upon Reason’s concept of latent and active failures, Human Factors Analysis and Classification System (HFACS) describes four levels of failure in the context of aircraft accidents: 1) Unsafe Acts, 2) Preconditions for Unsafe Acts, 3) Unsafe Supervision, and 4) Organizational Influences. Various sub components of these are given as

Categories of unsafe acts committed by aircrews

ERRORS

Skillbased Errors

• Breakdown in visual scan
• Failed to prioritize attention
• Inadvertent use of flight controls
• Omitted step in procedure
• Omitted checklist item
• Poor technique
• Overcontrolled the aircraft

Decision Errors

• Improper procedure
• Misdiagnosed emergency
• Wrong response to emergency
• Exceeded ability
• Inappropriate maneuver
• Poor decision

Perceptual Errors (due to)

• Misjudged distance/altitude/airspeed
• Spatial disorientation
• Visual illusion

VIOLATIONS

• Failed to adhere to brief
• Failed to use the radar altimeter
• Flew an unauthorized approach
• Violated training rules
• Flew an overaggressive maneuver
• Failed to properly prepare for the flight
• Briefed unauthorized flight
• Not current/qualified for the mission
• Intentionally exceeded the limits of the aircraft
• Continued low-altitude
  flight in VMC
• Unauthorized low-altitude
  canyon running 

Categories of preconditions of unsafe acts.

Substandard Conditions of Operators

Adverse Mental States

• Channelized attention
• Complacency
• Distraction
• Mental Fatigue
• Get-home-itis
• Haste
• Loss of situational awareness
• Misplaced motivation
• Task saturation

Adverse Physiological States

• Impaired physiological state
• Medical illness
• Physiological incapacitation
• Physical fatigue

Physical/Mental Limitation

• Insufficient reaction time
• Visual limitation
• Incompatible intelligence/aptitude
• Incompatible physical capability

Substandard Practice of Operators

Crew Resource Management

• Failed to back-up
• Failed to communicate/coordinate
• Failed to conduct adequate brief
• Failed to use all available resources
• Failure of leadership
• Misinterpretation of traffic calls

Personal Readiness

• Excessive physical training
• Self-medicating
• Violation of crew rest requirement
• Violation of bottle-to-throttle requirement

Categories of unsafe supervision.

Inadequate Supervision

• Failed to provide guidance
• Failed to provide operational doctrine
• Failed to provide oversight
• Failed to provide training
• Failed to track qualifications
• Failed to track performance

Planned Inappropriate Operations

• Failed to provide correct data
• Failed to provide adequate brief time
• Improper manning
• Mission not in accordance with rules/regulations
• Provided inadequate opportunity for crew rest

Failed to Correct a Known Problem

• Failed to correct document in error
• Failed to identify an at-risk aviator
• Failed to initiate corrective action
• Failed to report unsafe tendencies

Supervisory Violations

• Authorized unnecessary hazard
• Failed to enforce rules and regulations
• Authorized unqualified crew for flight

Organizational factors influencing accidents

Resource Management

• Human Resources
  • Selection
  • Staffing / manning
  • Training
• Monetary / budget resources
  • Excessive cost cutting
  • Lack of funding
• Equipment / facility resources
  • Poor design
  • Purchasing of unsuitable equipment

Organizational Climate

• Structure
  • Chain-of-command
  • Delegation of authority
  • Communication
  • Formal accountability for actions
• Policies
  • Hiring and firing
  • Promotion
  • Drugs and alcohol
• Culture
  • Norms and rules
  • Values and beliefs
  • Organizational justice

Organizational Process

• Operations
  • Operational tempo
  • Time pressure
  • Production quotas
  • Incentives
  • Measurement / appraisal
  • Schedules
  • Deficient planning
• Procedures
  • Standards
  • Clearly defined objectives
  • Documentation
  • Instructions
• Oversight
  • Risk management
  • Safety programs