ArticlePDF Available

Categorization of Near-Collision Close Calls Reported to the Aviation Safety Reporting System

Authors:

Abstract and Figures

Close calls in aviation are frequent occurrences. Many reports have described encounters with unmanned aerial vehicles or conflicts due to increasingly congested airspace. The Aviation Safety Reporting System, developed by the National Aeronautics and Space Administration, contains many close call narratives. However, few theoretical frameworks facilitate analyses of close call reports. This article describes an examination of close calls within the ASRS over a three-year period (2014-2016). The authors analyzed 117 close call reports from the ASRS involving near-midair collisions. Using Bliss’s (2013) taxonomy, the reports were successfully analyzed in terms of category and severity level. Results indicated that nearly half of the close calls occurred during the approach phase of flight. Also, a disproportionate number of close calls were “Un-signaled” and “Event-Driven.” Report frequency was negatively associated with aircraft separation distance. Recommendations include modification of the close call taxonomy to account for events caused by lack of responding.
Content may be subject to copyright.
Categorization of Near-Collision Close Calls Reported to the
Aviation Safety Reporting System
Lauren N. Tiller and James P. Bliss
Old Dominion University, Norfolk, VA
Close calls in aviation are frequent occurrences. Many reports have described encounters with unmanned
aerial vehicles or conflicts due to increasingly congested airspace. The Aviation Safety Reporting System,
developed by the National Aeronautics and Space Administration, contains many close call narratives.
However, few theoretical frameworks facilitate analyses of close call reports. This article describes an
examination of close calls within the ASRS over a three-year period (2014-2016). The authors analyzed
117 close call reports from the ASRS involving near-midair collisions. Using Bliss’s (2013) taxonomy, the
reports were successfully analyzed in terms of category and severity level. Results indicated that nearly
half of the close calls occurred during the approach phase of flight. Also, a disproportionate number of
close calls were “Un-signaled” and “Event-Driven.” Report frequency was negatively associated with
aircraft separation distance. Recommendations include modification of the close call taxonomy to account
for events caused by lack of responding.
INTRODUCTION
Close calls (also referred to as “near misses”) are
ubiquitous across task domains. The World Health
Organization (WHO) defines a near miss as a ‘‘serious error
or mishap that has the potential to cause an adverse event, but
fails to do so by chance or because it was intercepted” (World
Health Organization, 2005, p. 9). Medical practitioners,
surface transportation specialists, and the Federal Aviation
Administration have long understood the importance of
recording close call events. Frequently, investigation of such
events can help isolate necessary modifications to personnel
training, air traffic control or piloting procedures, or
equipment specifications.
In recent months, the media has presented several high-
profile close call events in aviation. For example, actor
Harrison Ford was involved in a “potentially serious incident”
on February 13, 2017, when he mistakenly landed his private
plane on a taxiway instead of his assigned runway at John
Wayne Airport (Costello & O'Hara, 2017). During his descent,
he experienced a near-collision with an American Airlines 737
plane on a taxiway that was carrying 110 passengers and six
crew members (Costello & O'Hara, 2017).
The proliferation of unmanned aircraft operating near
airports has resulted in increasing numbers of close calls as
well. As reported by NBC news, pilots reported 1,218 close
calls with drones in 2015. The next year, the number had
increased to more than 1,800 close calls (NBC News, 2017).
Bliss, Rice, Hunt, and Geels (2013) published a
taxonomy of close call terms that drew distinctions among
several closely related terms. They suggested that close call
events may represent a variety of possible scenarios that may
not be adequately explained by the Signal Detection Theory
framework. They proposed a specific categorization scheme
for classifying close calls that may aid in distinguishing
among six types of close call situations (un-signaled, event-
driven, response-driven, signaled, disregarded, and vicarious).
Because Harrison Ford made an incorrect response to land on
a taxiway (Costello & O'Hara, 2017) that event would be
categorized as a response-driven close call.
According to the Federal Aviation Administration, at any
given time there are 7000 aircrafts in the sky (Federal
Aviation Administration, 2016). Actions by aviators and flight
crews are generally predictable based on Air Traffic Control
directives, mandates from the FAA, directives from flight
carriers, and actions of other aircraft operators. Research has
shown that unless the high probabilities of risks are known,
current routines are unlikely to change unless accidents or
incidents have occurred (Shimazoe & Burton, 2013).
The availability of close call data may provide an
alternative to relying on accidents to drive change. Maintained
within databases such as the Aviation Safety Reporting
System, close call reports are abundant with important
information about the issues that arise during human
technology interactions. Aviation close call reports also
provide information needed to categorize the occurrence by
the “type” of close call that took place. If enough information
is gathered about a certain type of close call, it can be used for
educational purposes to make the high probability of a risk
known before similar instances become accidents. Research
has acknowledged that in the mind of the individual human
decision maker, close calls are usually deemed as “successes”
and the near misses reinforce the individual’s current routine.
When a “justification shift” occurs, it leads to an
underestimation of the known risk (or close call) and an
overestimation of the reliability of the routine (Shimazoe &
Burton, 2013). To better communicate these risks, it is
essential to first understand the differences that exist among
the types of close call events.
It is equally important to classify close calls by the
severity of event. As highlighted by Bliss et al. (2013), if a
risk is quantified effectively, certain close calls can be
weighted more heavily than others for training purposes. The
severity of a close call can be determined by assessing the
consequences that could have happened if the close call had
materialized into actual danger (Bliss et al., 2013). Dillion,
Tinsley, and Burns (2014) studied how close call perception
Copyright 2017 by Human Factors and Ergonomics Society. DOI 10.1177/1541931213601947
Proceedings of the Human Factors and Ergonomics Society 2017 Annual Meeting 1866
alters future preparation behavior, classifying near miss events
as either a “resilient near misses” or “vulnerable near misses”.
They defined a resilient near miss as an occurrence wherein
the individual sees that negative consequences did not occur
and underestimates future danger. In vulnerable near miss
instances, the individual regards avoidance of danger as a
consequence of luck; therefore, no future action is warranted
(Dillion et al., 2014). Their study results revealed a fully
mediated model, wherein the more an event was considered to
resemble a vulnerable near miss, the higher the perceived risk,
and the higher the intent to take future preventative action
(Dillion et al., 2014).
Dillion et al.’s model makes an important contribution to
the study of close calls from the operator’s perspective.
However, it provides little direction about how to effectively
categorize close call events to quantify severity or to
determine whether the close call information is actionable for
increases in safety.
FOCUS OF THIS INVESTIGATION
For the current research, the authors studied the
occurrences of aviation close calls within the ASRS database.
Individual situation reports were interpreted and categorized
according to Bliss et al.’s (2013) taxonomy. Key objectives
were to investigate operators’ reactions and the human-
technology interactions that occurred. The ultimate purpose
was to establish the utility of Bliss et al.’s taxonomy for
clarifying differences among close call events and for
determining whether severity of reported close call events
could be meaningfully classified in terms of reported
separation distance.
Method
The Aviation Safety Reporting System (ASRS) Database
Online. The ASRS database is a repository of voluntarily
submitted event reports created by the National Aeronautics
and Space Administration (NASA) for use in research
pertaining to aviation safety. The reports referring to safety
incidents are considered soft data and are subject to self-
reporting bias. Reports may be downloaded free of charge
from https://asrs.arc.nasa.gov/search/database.html.
Search Procedure. Reports were downloaded using
ASRS online database from the NASA website on January 30,
2017. The authors specifically analyzed ASRS anomaly
conflicts that pertained to near mid-air collisions (NMAC) and
critical airborne or ground conflicts.
All reports in the database identified more than one
aircraft. For the purposes of this investigation, our analyses
were conducted from the first aircraft’s perspective. Of the
15,438 reports submitted to the ASRS in 2014, 2015, and
2016, 133 reports met the specific criteria we used to isolate
close call NMACs. The resulting reports involved all
operation types except training. The specific filtering method
used is described in the following paragraph.
Using the “Aircraft Mission” filter within the ASRS
query tool, we commanded the system to return all types of
operations except training. Using the “Event Type” filter, we
searched all subcategories but excluded published material or
policies and security issues from the “Procedural Deviation”
subcategory. From the subcategory “Flight
Deck/Cabin/Aircraft” we excluded illness, passenger
electronic devices, and passenger misconduct. Using the event
assessment detector filter, we included all subcategories but
excluded gate agents/customer service representatives,
observers, and passengers from the “Person Detector”
subcategory. Using the “Human Factors” filter, we used all
event subcategories except communication breakdown and
training/ qualification. Finally, the text filter was used to
search the synopsis and narrative for text that contained “Near
Miss”.
Using all of the above search parameters simultaneously,
the database returned 133 results (39 in 2014, 49 in 2015, and
45 in 2016). This number was later reduced to 117 after
removing reports that contained inadequate information to
determine the type of close call category.
Before analyzing the reports, a severity taxonomy was
created according to recommendations within Bliss (2013).
Basic flight rules suggest that aircraft separation must be at
least 1000 ft vertically or horizontally for Instrument Flight
Rules (IFR) traffic, or 500 ft for Visual Flight Rules (VFR)
traffic (Kayton & Fried, 1997). From these values, the authors
constructed nine severity categories, as detailed within Table 1
below. Our intent was to develop a taxonomy that was
objective and generalizable across conditions and aircraft.
Table 1. System for Assignment of Close Call Severity.
NMAC
Close Call
Severity
Aircraft
Separation
Distance
L9 </= 100 ft
L8 101 – 200 ft
L7 201 – 300 ft
L6 301 – 400 ft
L5 401 – 500 ft
L4 501 – 600 ft
L3 601 – 700 ft
L2 701 – 800 ft
L1 >/= 801 ft
Absolute aircraft separation distance was calculated by
using the reported vertical and lateral miss distances and
implementing the Pythagorean Theorem to find the
hypotenuse leg for each report.
After retrieving the reports, each was classified by the
type or types of close calls that occurred during an event. The
types of close call categories used were Signaled Close Calls
(SCC), Un-Signaled Close Calls (UCC), Event-Driven Close
Calls (ECC), and Response-Driven Close Calls (RCC)(Bliss et
al., 2013).
Each ASRS report narrative was analyzed by isolating
the following specific information: phase of flight for all
aircrafts involved, type of anomaly conflict (NMAC, Critical
Airborne Conflict, or Ground Conflict), and the detector(s) of
the near-miss (flight crew member, air traffic controller, or
Proceedings of the Human Factors and Ergonomics Society 2017 Annual Meeting 1867
onboard Traffic Collision Avoidance System). We
documented the calculated miss distance, saliency level of
severity, and category or categories of the close calls that
occurred during an event. A subjective decision was reached
by the authors (one of whom has had flight training) regarding
whether the pilot could have executed a different plan of
action that would have avoided the near miss event.
Every event was classified as either a Signaled Close
Call (SCC) or Un-Signaled Close Call (UCC). If an event was
classified as a SCC, the collision danger was anticipated by
either a mechanical signal or by an air traffic controller.
Conversely, if the event was classified as a UCC, the collision
danger was anticipated by a member of the flight crew in the
first aircraft. In some cases, the collision danger was not
anticipated by anyone or anything. Some narratives were
interpreted as Event-Driven Close Calls (ECC’s) or Response-
Driven Close Calls (RCC’s). Event-Driven Close Calls
(ECC’s) included situations that evolved into close calls
without interruption or intervention by the flight crew of the
first aircraft. This may have occurred because there was
inadequate time for the flight crew to complete an evasive
maneuver or otherwise correct the situation. Response-Driven
Close Calls (RCC) occurred when the situation evolved into a
close call because of something that was specifically done or
not done by the flight crew of the first aircraft. For example,
increasing the aircraft’s altitude may have brought the aircraft
into closer proximity with a second aircraft. Notably, close
calls that materialized because of failure to respond when a
response was warranted were also categorized as RCC’s.
RESULTS
From the narrative accounts within the ASRS, it was
relatively straightforward to determine how the impending
collision was initially detected. In most cases (84 of 117), the
Pilot Flying or Pilot Monitoring (Pilot Not Flying) observed
the impending collision. In 17 cases, the impending collision
was observed by an air traffic controller.
Of the 117 close call near-collision reports investigated
(including near-midair collisions and near-collisions on the
ground), only 22 narratives made specific mention of Traffic
Collision Avoidance System (TCAS) activity. This may be
due, in part, to the variety of aircraft included within the
reports. For example, gliders are not required to have an
operational TCAS system on board. In at least two cases,
however, the report specifically mentioned that TCAS failed
to notice traffic in the area. This often happens when drones
are the source of the near mid-air collision. Another
consideration is that TCAS must be operational on both
aircraft for a traffic advisory or resolution advisory to be
generated. This was not always the case, especially if one of
the aircraft was a private airplane.
In Table 2 below, the close call narratives are separated
into taxonomic categories according to the descriptions of
such categories within Bliss et al.’s (2013) article. Also
included are the number of close calls in each taxonomic
category that were interpreted as actionable (pilot action could
or did resolve the situation) or non-actionable (situation
resolution was outside of the pilot’s control). Because the
ASRS includes personal reports, none of the reports were
interpreted as close calls that were vicarious or disregarded by
flight crew members.
Table 2. Actionable and Non-Actionable Close Calls by
Taxonomy Category.
Actionable Non-
Actionable
Total
Disregarded 0 0 0
Event-Driven 18 38 56
Response-Driven 18 0 18
Signaled 18 6 24
Un-signaled 40 44 84
Vicarious 0 0 0
Figure 1 shows the frequency of close call events by the
specific phase of flight. The majority of near mid-air
collisions occurred during the approach phase, as aircraft were
nearing an airport or were entering a landing flight pattern. It
is appropriate to combine the “initial approach,” “descent” and
“final approach” phases. Doing so reveals that 56 of the 117
(48%) close calls occurred as aircrafts were preparing to land.
Relatively few close calls were reported during the departure
and landing phases (see Figure 1).
Close calls by Phase of Flight
Figure 1. Frequency of Close Calls by Phase of Flight.
Figure 2 shows the frequency of close calls observed by
level of severity (L1 – L9). Across all events reported, the
average distance between aircrafts was 265 feet. Specific
values ranged from 2 feet (an outlier reflecting a ground
taxiway close call) to 2,121 feet. As shown in Figure 2, there
is an increasing, monotonic relationship, with greater numbers
of close calls associated with greater severity. This may reflect
individual reporting bias. Some flight crew members may not
consider distances among aircraft greater than 500 feet to be a
“close call;” however, it is likely that any flight crew member
would qualify closer distance separations as close calls.
0
5
10
15
20
25
30
Number of Close Calls
Proceedings of the Human Factors and Ergonomics Society 2017 Annual Meeting 1868
Figure 2. Frequency of Close Calls by Level of Severity.
DISCUSSION
The data reported here represent an attempt to validate
the taxonomic structure advocated by Bliss et al. (2013). The
authors have suggested that close calls may fall into a number
of categories depending on whether or not they were preceded
by some sort of signal (Signaled or Un-Signaled), whether
they occurred independent of an operator’s actions or because
of them (Event-Driven or Response-Driven), and whether they
failed to command operator attention because they were
irrelevant (Disregarded) or they happened to someone else
(Vicarious).
Generally, the classification system appears to have been
successful for analyzing close call types. The categories
described above clearly distinguished the 117 narratives,
pointing out useful trends within those reports. However, we
did note that the “Response-Driven” close call category lacks
granularity to separate close calls that resulted from operator
action from those that resulted from operator inaction. It may
be useful to incorporate an additional category to refer to the
latter circumstance.
Our current consideration of close calls in aviation
shows that a large percentage (72%) of events reported were
not preceded by any sort of signal. Rather, flight crew
members encountered other traffic unexpectedly. The
possible reasons for this vary. Frequently, small (general
aviation) aircraft do not have a traffic collision system on
board, whereas commercial aircraft generally do. In other
cases, available signals from air traffic controllers or collision
avoidance warning systems may have referred to multiple
threats.
Within the narratives themselves, authors frequently
mentioned the need for TCAS systems to be sensitive to drone
traffic. Typically, drones are small enough to be missed
entirely by air traffic control or TCAS. As noted recently by
NBC News (2017), the FAA has begun to examine detection
systems for drones, but the earliest systems may not be
available for several years.
The data reported here also highlight a fact that has been
known for many years: the potential for near midair collisions
is greatest around airports. In fact, our analysis of the ASRS
narratives shows that nearly half of the events occurred as
airplanes were approaching airports to land. At very busy
airports, the problem may be compounded by the existence of
parallel approaches, where aircraft may be landing in close
proximity to each other. As the FAA moves toward
compressed schedules that accompany implementation of the
new “NextGen” airspace system, the problem of destination
airport congestion may become even worse.
In addition to collision system detection enhancement,
another essential use of close call databases is the potential for
training improvements. For example, our current investigation
underscores the need for continued training of pilots, air traffic
controllers, and flight crew personnel in a variety of areas. As
noted, the TCAS system currently fails to detect unmanned
aircraft as well as other aircraft that do not have operational
TCAS on board. For that reason, it is important that aviators
be encouraged and trained to maintain good out-the-window
scanning behavior as well as scanning of radio traffic.
As noted above, close calls remain a pervasive
occurrence across task domains. Toward that end, it is
important that the validation effort reported here be replicated
using data within other report databases. Candidate task
domains include firefighting and medical operations. It is also
necessary to conduct empirical research to further demonstrate
the causative link between close call types and subsequent
operator behavior.
ACKNOWLEDGMENT
The authors acknowledge the assistance of CPT John M. Bliss,
who provided edits to the article.
REFERENCES
Federal Aviation Administration (FAA)(2016). Air traffic by
the numbers. Online. Retrieved March 7, 2017, from
Federal Aviation Administration:
https://www.faa.gov/air_traffic/by_the_numbers/
Bliss, J. P., Rice, S., Hunt, G., & Geels, K. (2013). What are
close calls? A proposed taxonomy to inform risk
communication research. Saftey Science 61 (2014),
21-28.
Costello, T., & O'Hara, M. E. (2017, February 14). Harrison
Ford in Incident with passenger plane at California
airport. Retrieved March 7, 2017, from NBC News:
http://www.nbcnews.com/nightly-news/harrison-
ford-has-incident-passenger-plane-airport-n720826
Dillion, R. L., Tinsley, C. H., & Burns, W. J. (2014). Near-
misses and future disaster preparedness. Risk
Analysis, 34, 1907-1922.
Kayton, M., & Fried, W. R. (1997). Avionics Navigation
Systems 2nd edition. New York, NY: John Wiley &
Sons, Inc.
National Aeronautics and Space Administration
(NASA)(2011). Accident Precursor Analysis
Handbook. Retrieved March 2, 2017, from
https://www.hq.nasa.gov/office/codeq/doctree/NASA
_SP-2011-3423.pdf
0
5
10
15
20
25
30
35
40
Number of Close Calls
Severity Level
Proceedings of the Human Factors and Ergonomics Society 2017 Annual Meeting 1869
NBC News (2017). A look at the dangerously close
encounters between drones and planes. Retrieved
February 10, 2017 from
http://www.nbcnews.com/nightly-news/video/a-look-
at-the-dangerously-close-encounters-between-drones-
and-planes-874783811506
Shimazoe, J., & Burton, R. M. (2013). Justification shift and
uncertainty: why are low-probability near misses
underrated against organizational routines?
Computational Mathmatical Organizational Theory,
19, 78-100.
World Health Organization (2005). WHO Draft Guidelines for
Adverse Event Reporting and Learning Systems.
Author, Geneva, Switzerland.
Proceedings of the Human Factors and Ergonomics Society 2017 Annual Meeting 1870
... Firefighting is not unique with regard to close calls. Close calls are a frequent occurrence in many complex task situations, such as aviation [7], medicine, surface transportation, construction [8] and even sports [9]. The generality of the close call phenomenon has rendered it difficult to study, though some researchers have attempted to construct theoretical models to enable prediction, exploitation and mitigation of close call events [10]. ...
... The current research represents part of an ongoing effort to apply the taxonomic structure advocated within Table 1 to a series of applied task domains. Recently, Tiller and Bliss [7] addressed this goal by conducting an examination of the Aviation Safety Reporting System (ASRS). They demonstrated the relevance of the close call categories described in Table 1 to a collection of extracted ASRS incident reports. ...
... The findings reported here constitute an important bridge between theoretical treatments of close calls and applied domains within which they frequently occur. Tiller and Bliss [7] noted that close call events within the Aviation Safety Reporting System followed the categories established in Bliss's [11] taxonomy. Similarly, the events noted within the IAFC database seemed well represented by the categories, with one exception (Disregarded Close Calls). ...
Chapter
Close calls or near misses are common occurrences in the field of firefighting. As in many other domains, they represent a source of valuable information to inform personnel training and improve safety. Firefighting is uniquely complex because of the many possible causes of injury or death. Database repositories for close call narrative reports exist; however, they contain a wide variety of information that varies in usefulness. The authors applied a published taxonomy to categorize narrative reports within the www.firefighternearmiss.com online database. The authors analyzed 61 narratives from the database, categorizing them by situational variables, level of situation severity, and reported error frequencies. Results showed that the majority of narratives described close call events that were preceded by a sensory signal and attributable to human error. Notably, the vast majority of narratives included information from which actionable recommendations could be generated. The article concludes with suggestions for future database development.
... A near-miss is "where a negative outcome could have happened because of hazardous conditions but did not" (Dillon, Tinsley, & Burns, 2014, p. 1980. Near-misses such as nearcollisions between drones and passenger jets are commonly reported in media (Tiller & Bliss, 2017), which results in negative public attention. While near-miss is rarely (if at all) studied by public relations researchers, paracrisis is a closely related concept. ...
... This study argues that reduced trust could be a negative outcome associated with nearmiss reporting. Near-misses between airborne vehicles are often reported by news media (Tiller & Bliss, 2017). For example, over the last year, online news reported near-miss between passenger jets and aircraft in the United Kingdom (Loganair flight in 'near miss' with drone near Glasgow Airport, 2019, 15 February; Near-miss between passenger jet & drone over London, report reveals, 2018, 21 October). ...
Article
Full-text available
A near-miss, such as a narrowly avoided collision between vehicles, evades a full-scale accident but can generate media attention that threatens trust. In emerging industries, the effects of near-miss may extend beyond organizational boundaries and influence trust in the industry and technology. This study empirically tests these assertions by examining how media reports of near-miss affect organizational trustworthiness and how trust repair strategies after a near-miss influence organizational and industry trust and technology acceptance in the emerging commercial industry of unmanned aircraft. Notwithstanding parallels to paracrisis, near-miss communication is understudied in public relations research. Further, studies of trust in the context of crisis are recent (see Brühl et al., 2018; Fuoli et al., 2017), and have produced unexpected results that warrant continued exploration in public relations. Underpinned by attribution theory, this study adopts a 2 (near-miss cause: external, low controllability; internal, high controllability) x 3 (trust repair strategy delivered via news story: denial, excuse, apology) scenario-based experiment. This study found that near-miss reduced organizational trustworthiness regardless of whether the event was controllable or not, indicating that when it comes to trust perceptions, near-miss can operate similarly to crisis. Further, apology was the only strategy that arrested a fall in organizational trustworthiness. The study signaled a trust transfer effect where organizational trust influenced industry trust, which led to the acceptance of unmanned aircraft technology. In the context of emerging industries, these findings have implications for organizations that experience near-miss, highlighting the potential for a standardized initial strategy to acknowledge a reduction in trust in order to support trust beyond the organization.
Article
Full-text available
Aviation is a complicated transportation system, and safety is of paramount importance because aircraft failure often involves casualties. Prevention is clearly the best strategy for aviation transportation safety. Learning from past incident data to prevent potential accidents from happening has proved to be a successful approach. To prevent potential safety hazards and make effective prevention plans, aviation safety experts identify primary and contributing factors from incident reports. However, safety experts’ review processes have become prohibitively expensive nowadays. The number of incident reports is increasing rapidly due to the acceleration of advances in information technologies and the growth of the commercial and private aviation transportation industries. Consequently, advanced text mining algorithms should be applied to help aviation safety experts facilitate the process of incident data extraction. This paper focuses on constructing deep-learning-based models to identify causal factors from incident reports. First, we prepare the data sets used for training, validation, and testing with approximately 200,000 qualified incident reports from the Aviation Safety Reporting System (ASRS). Then, we take an open-source natural language model, which is well trained with a large corpus of Wikipedia texts, as the baseline and fine-tune it with the texts in incident reports to make it more suited to our specific research task. Finally, we build and train an attention-based long short-term memory (LSTM) model to identify primary and contributing factors in each incident report. The solution we propose has multilabel capability and is automated and customizable, and it is more accurate and adaptable than traditional machine learning methods in extant research. This novel application of deep learning algorithms to the incident reporting system can efficiently improve aviation safety.
Article
This study examines how exploiting biases in probability judgment can enhance deterrence using a fixed allocation of defensive resources. We investigate attacker anchoring heuristics for conjunctive events with missing information to distort attacker estimates of success for targets with equal defensive resources. We designed and conducted a behavioral experiment functioning as an analog cyber attack with multiple targets requiring three stages of attack to successfully acquire a target. Each stage is associated with a probability of successfully attacking a layer of defense, reflecting the allocation of resources for each layer. There are four types of targets that have nearly equal likelihood of being successfully attacked, including one type with equally distributed success probabilities over every layer and three types with success probabilities that are concentrated to be lowest in the first, second, or third layer. Players are incentivized by a payoff system that offers a reward for successfully attacked targets and a penalty for failed attacks. We collected data from a total of 1,600 separate target selections from 80 players and discovered that the target type with the lowest probability of success on the first layer was least preferred among attackers, providing the greatest deterrent. Targets with equally distributed success probabilities across layers were the next least preferred among attackers, indicating greater deterrence for uniform‐layered defenses compared to defenses that are concentrated at the inner (second or third) levels. This finding is consistent with both attacker anchoring and ambiguity biases and an interpretation of failed attacks as near misses.
Article
Full-text available
Knowledge of near misses is helpful in preventing accidents, but it does not always lead to changes in organizational routines or contribute to accident prevention. In this article, the authors argue that low-probability near misses reinforce beliefs of experts and professionals in existing routines, proposing a concept of justification shift. Justification shift is underestimation of risks of known near misses vis-à-vis overestimation of reliabilities of existing routines. Consequently, signals of “close calls” tend to be disregarded, and accidents become more likely. When justification shift occurs, experts and professionals who want changes in routines face burden of proof about risks of those routines. Uncertainty in requirements and data may increase the burden and make justification shift more likely. To explore how justification shift occurs and how uncertainty influences the shift, the authors have developed an agent-based model based on theories in organizational studies and the case of the space shuttle Challenger accident in 1986. From the simulation using the model, it becomes clear that uncertainty does not make a difference in frequencies and degrees of justification shift. The authors also discuss implications of this finding to efforts to utilize knowledge of near misses and to further research on the efforts and organizational decisions.
Article
Disasters garner attention when they occur, and organizations commonly extract valuable lessons from visible failures, adopting new behaviors in response. For example, the United States saw numerous security policy changes following the September 11 terrorist attacks and emergency management and shelter policy changes following Hurricane Katrina. But what about those events that occur that fall short of disaster? Research that examines prior hazard experience shows that this experience can be a mixed blessing. Prior experience can stimulate protective measures, but sometimes prior experience can deceive people into feeling an unwarranted sense of safety. This research focuses on how people interpret near-miss experiences. We demonstrate that when near-misses are interpreted as disasters that did not occur and thus provide the perception that the system is resilient to the hazard, people illegitimately underestimate the danger of subsequent hazardous situations and make riskier decisions. On the other hand, if near-misses can be recognized and interpreted as disasters that almost happened and thus provide the perception that the system is vulnerable to the hazard, this will counter the basic “near-miss” effect and encourage mitigation. In this article, we use these distinctions between resilient and vulnerable near-misses to examine how people come to define an event as either a resilient or vulnerable near-miss, as well as how this interpretation influences their perceptions of risk and their future preparedness behavior. Our contribution is in highlighting the critical role that people's interpretation of the prior experience has on their subsequent behavior and in measuring what shapes this interpretation.
Book
The Navigation Equations (M. Kayton). Multisensor Navigation Systems (J. Huddle & R. Brown). Terrestrial Radio-Navigation Systems (B. Uttam, et al.). Satellite Radio Navigation (A. Van Dierendonck). Terrestrial Integrated Radio Communication-Navigation Systems (W. Fried, et al.). Inertial Navigation (D. Tazartes, et al.). Air-Data Systems (S. Osder). Attitude and Heading References (M. Kayton & W. Wing). Doppler and Altimeter Radars (W. Fried, et al.). Mapping and Multimode Radars (J. Pearson, et al.). Celestial Navigation (E. Knobbe & G. Haas). Landing Systems (D. Vickers, et al.). Air Traffic Management (C. Miller & J. Scardina). Avionics Interfaces (C. Spitzer). References. Index.
Air traffic by the numbers, from Federal Aviation Administration: https://www.faa.gov/air_traffic What are close calls? A proposed taxonomy to inform risk communication research
  • J P Bliss
  • S Rice
  • G Hunt
  • K Geels
Federal Aviation Administration (FAA)(2016). Air traffic by the numbers. Online. Retrieved March 7, 2017, from Federal Aviation Administration: https://www.faa.gov/air_traffic/by_the_numbers/ Bliss, J. P., Rice, S., Hunt, G., & Geels, K. (2013). What are close calls? A proposed taxonomy to inform risk communication research. Saftey Science 61 (2014), 21-28.
Harrison Ford in Incident with passenger plane at California airport
  • T Costello
  • M E Hara
Costello, T., & O'Hara, M. E. (2017, February 14). Harrison Ford in Incident with passenger plane at California airport. Retrieved March 7, 2017, from NBC News: http://www.nbcnews.com/nightly-news/harrisonford-has-incident-passenger-plane-airport-n720826
A look at the dangerously close encounters between drones and planes
  • M Kayton
  • W R Fried
Kayton, M., & Fried, W. R. (1997). Avionics Navigation Systems 2nd edition. New York, NY: John Wiley & Sons, Inc. National Aeronautics and Space Administration (NASA)(2011). Accident Precursor Analysis Handbook. Retrieved March 2, 2017, from https://www.hq.nasa.gov/office/codeq/doctree/NASA _SP-2011-3423.pdf NBC News (2017). A look at the dangerously close encounters between drones and planes. Retrieved February 10, 2017 from http://www.nbcnews.com/nightly-news/video/a-lookat-the-dangerously-close-encounters-between-dronesand-planes-874783811506
WHO Draft Guidelines for Adverse Event Reporting and Learning Systems. Author
World Health Organization (2005). WHO Draft Guidelines for Adverse Event Reporting and Learning Systems. Author, Geneva, Switzerland.
A look at the dangerously close encounters between drones and planes
  • Nbc News