Data Fusion

 Data Fusion

By Catalina9

Data fusion is the process of integrating multiple data sources to produce more consistent, accurate, and useful information than what is provided by any individual data source. Data fusion is the process of getting data from multiple sources in order to build more sophisticated models and understand more about a project. It often means getting combined data on a single subject and combining it for central analysis. In airport operations data fusion is to obtain reports from multiple sources such as weather information, NOTAM, hazards, incidents or accidents and analyze in a single source Statistical Process Control (SPC). Data fusion is to analyze how a change in one source affects, or trickle down to the other sources within a system. A Safety Management System is a Data Fusion System.

Data Fusion is the data comprehension process.

Data supporting an SMS is always available but needs to be captured before being put into use. Capturing data is more than receiving reports, it is also to conduct research to extract information. Just data in itself is not an asset to a safety management system, but within a Data Fusion System it becomes an invaluable asset. Data collected must first be transformed into information. When information is consumed, through any of the five senses, information becomes knowledge. With knowledge comes comprehension of one, two or more systems within a Data Fusion System. 

SMS is a businesslike approach to safety. In a business data equals cash, which always is there, but unless captured it is not a source to be used in operations. Data to an SMS Enterprise is just as much value to operations as what cash is to a business. During the pre-SMS days, accident investigation focused on pilot errors, or what task a person failed to do. At some point after the 1956 Grand Canyon disaster, problem solvers decided that the pilot, or a human was the problem. In 1956 when two aircraft crashed over the Grand Canyon, the probable cause was not determined to be what the pilots failed to do, but that the pilots did not see each other in time to avoid the collision du to intervening clouds, visual limitations due to cockpit visibility, preoccupation with normal cockpit duties, preoccupation with matters unrelated to cockpit duties such as attempting to provide the passengers with a more scenic view of the Grand Canyon area, physiological limits to human vision and insufficiency of enroute air traffic advisory information. Probable causes for the disaster included human factors, organizational factors, supervision factors and environmental factors. The probable cause was based in data fusion and how data from multiple sources trickles down to one outcome.  

Data Fusion is where data become available
without subjective random selection

For data fusion to functioning as intended, data needs to be tied to an open-source project. Within the SMS an open-source project is where data is freely available, it is shared, and it is decentralized. No single enterprise, being large, medium, or small, has enough data available within its own organization to apply a data fusion system in their operations. Data fusion is in essence to reverse the reactive data collection process when data, or hazards must be visible for collection, to applying a proactive collection process where data available is trickling into multiple systems.       

Data collection is a major task to operate with an effective SMS. There is an expectation that the aviation industry collects its own data and apply that data to their operations. The aviation industry also expects that data collected by one enterprise is not applicable, or directly shareable, with another enterprise within the same industry. SMS guidance material states that “Your organization must develop policies, processes and procedures that support your unique operating requirements and that fit the size and complexity of your organization.” When SMS first was implemented by regulations, these types of statements were incorrectly interpreted that their own data would only be applicable to their own organization and that nobody else should benefit from their discoveries. 

The same misconception was for the Accountable Executive (AE), that their only role was to have control of the financial and human resources that were necessary for their activities and operations. When in fact, the role and responsibilities of an AE is to be accountable on behalf of an enterprise to meet all the requirements of the regulations. In other words, an AE must not only have knowledge of all regulations for compliance, but also know and comprehend what processes are necessary for compliance and what processes drifts towards non-compliance. An AE without a confidential adviser stands no chance to maintain this compliance requirement.

All data for an SMS system is already out there, but needs to be captured, filtered, and analyzed. Capturing data comes in many shapes and forms. Just as cash in a business comes in bills, coins or plastic, data in an SMS comes as hazards, incidents, or emotions. Capturing cash is by affecting human behavior in purchasing patterns, expectations, and social acceptance. Capturing data for SMS is generally a reactive process and is not “cashed in” until after an event. The capture of SMS data will benefit greatly by learning from the advertising and marketing industry by affecting human behaviors to recognize their own patterns, expectations, and social acceptance within and SMS enterprise, or in other words, changing their organizational culture to a just-culture. Within a just-culture is when data fusion becomes available. Filtering data is not associated with subjective selection of data, but it is to filter data into their pre-established safety critical areas and safety critical functions. Filtered data is then analyzed in a Statistical Process Control (SPC) System, where emotions and subjective randomization are eliminated from the equation. When SPC is eliminated from the analysis with a reaction to a statistical trend, there is a tendency for overcontrol, and applying one or two events as a trend, which again leads to a greater hazard than the hazard itself.

Data Fusion is tracking, mapping and loop closure.

A Data Fusion System is a system where there is tracking, mapping and loop closure. It is also a system where data trickling from one system into another system affects both the system it is trickling from and the system it is trickling into, which also is defined as organizational drift. The tools to implement a Data Fusion System into SMS is by applying a data collection tool that your enterprise has full control over. It is to conduct research, development, project planning and work directed toward the innovation, introduction, and improvement of processes. Data Fusion is a condition, while the audit result is the symptom.  

 

Catalina9

SMS QA Control Management: Audits and Golfing

SMS QA Control Management: Audits and Golfing:   Audits and Golfing By Catalina9 A n audit of your Safety Management System (SMS) is not to find out what, when, where, why, who and how ev...

Audits and Golfing

 Audits and Golfing

By Catalina9

An audit of your Safety Management System (SMS) is not to find out what, when, where, why, who and how everything went wrong with your safety management system, but to confirm what was done right. As an additional, or parallel level of safety, a safety management system is a tool to identify hazards before they affect operations. A safe enterprise may attract more clients, generate a better return on investment and enjoy their success. Success stories should encourage other companies to put in place systems that meet the SMS regulations. However, implementing an SMS in development is a higher level of hazard than operating without a formal SMS since safety is assumed with the system in place, while there is a lack of audits to verify safety in operations.

An audit is the compass to maintain course

  

Audits is to verify what goes right or what makes operation safe and joyful ask a customer experience. SMS is a system that is required to mature in order to paint a picture of the operations itself. This holds true for both airport and flight operations. When the implementation of SMS is forced to be effective it takes on the face of a two-year old and opposes the enforcement. It is not that anyone does anything wrong or incorrect, but the good intent to improve safety cannot be forced onto the SMS system. SMS has to build its own model, which will vary from enterprise to enterprise. No two airports, airlines or persons are alike and that is the reason why two safety management systems will be alike. Just like in a chess game, no two players are alike, but still, both players may win the world championship. Simplified, running an SMS is like playing golf. 

Daily Quality Control is to stay on track.

An audit of golf performance is in the competition itself. The competition is not to compare golfing performance with other golfers, but to compare golfing to an expected performance. The person who wins the golf tournament may not be the person with the best golfing history, or safety history, but is won by the person with the most reliable process to analyse special cause variations. Wind, and wind direction is one major special cause variation in golfing. Misjudging the wind takes you out of the game, or in other words, causing an incident. There are several special cause variations in golfing. Special cause variations in golfing are wind, temperature, dewpoint, air pressure, elevation, precipitation type and intensity, cloud cover, geographical location or geoid undulation. Some of these may be considered common cause variations since they are all required and common parts of an environmental system. What makes these variations special cause variations are their variation in intensity, amount, and behavior. In addition, these special cause variations are interacting within a 3D environment, while the golfer is observing, or calculating from outside of the 3D. A 3D environment in golfing is to view the environment from the golf ball’s point of view, just as an SMS enterprise must view the environment from an affected person’s point of view.  

 

Golfing interacts with a conglomerate of systems. When teeing off the golfer must maintain control of the system, from where the toes are pointing to where the hands and fingers are and to the intensity, amount, and behavior of the swing. Golf balls are designed and manufactures within the allowable variations of a standard and comes with an aerodynamic design for reliability. The concept is that when two balls are hit exactly the same way they will arrive at exactly the same location. This principle may be true in a virtual world where locations are mathematically calculated. On the golf course the locations where balls arrive are statistical data collection and its variation from mathematical calculation. 

Daily quality control and audits are connected by the circles of infinity.

Some of the special cause variations may be estimated at the time of the swing, while other may be mathematically calculated to determine the intensity, amount, and behavior of the swing. Winds are estimated and the swing adjusted. Elevation, temperature and dewpoint are calculated, since their variation are gradual over time. Even a cold front temperature change within an hour is a gradual change. A wind gust is a variable within a split second, or an instant change. A geoid undulation is a change in distance between geographical coordinates with elevations, where the distance between geographical coordinates at sea level or at altitude could vary as much as a foot. Some of these special cause variations may be assessed as significant to performance but sill inconsequential to the end result since after the swing there is no turning back to alter the location where the golf ball will arrive. Each round of golf, or every golfing competition is its own audit of every candidate, or system attending. The quality control, which is a prerequisite for an audit, is in the data collection, testing and analysis of a golfer’s swing prior to the audit. 

 

An audit of an airport, airport or a safety management system follows the same process as golfing, with a daily quality control system, data collection, testing analysis of the systems. An audit of an SMS enterprise is in the operations itself, or how things go right, or as expected from a customer’s point of view. The weather is a significant variable in how well an enterprise performs, or how safe they operate. At an airport, a contaminated runway is a runway when a significant portion of the runway surface area is affected by compacted snow, dry snow, frost, ice, slush, standing water, wet ice, or wet snow. A daily quality control system is an oversight system where contamination is analyzed and accepted or rejected by the airport operator. The next step is for the airline, or pilot, to accept or reject the runway surface conditions. This decision in turn becomes the audit result to assess the safety in operations by an enterprise. The enterprise itself may have completed their required tasks, but they overlooked how their decision affected aircraft operations, which is turn is the audit result, or in other words the golf tournament result.  

 

An audit of a safety management system is an audit of a parallel system to the requirements of operations itself. An enterprise has a responsibility to do what it takes to ensure safety when the operational regulations are without a system to assess safety. Aircrews must assess the aerodrome as being suitable for the intended operation, while an airport operator is not required to make this assessment. However, prior to SMS there was nothing that informed an aircraft operator as to the airport level of service provided at the airport. Only a general statement was provided as to whether or not the runways were maintained in compliance with airport standards. This general statement did not provide the aircraft operator, or flight crew, adequate detail as to the suitability of each runway, taxiway, or apron at an airport. As an additional level of safety, or an oversight system parallel to the operations, an audit of an airport captures a painted picture of the end result and not a snapshot of regulatory compliance level. 

 

 

 

Catalina9