Presentation is loading. Please wait.

Presentation is loading. Please wait.

Improving Safety through Collaborative Safety Initiatives

Similar presentations


Presentation on theme: "Improving Safety through Collaborative Safety Initiatives"— Presentation transcript:

1 Improving Safety through Collaborative Safety Initiatives
Commercial Aviation Safety Team (CAST) Aviation Safety Information Analysis Sharing (ASIAS) Presented to ICAO Air Navigation Commission Jay Pardee Director, Office of Accident Investigation and Prevention October 6, 2009 1

2 Commercial Aviation Safety Team (CAST) Overview

3 In the U.S., our focus was set by the White House Commission on Aviation Safety, and The National Civil Aviation Review Commission (NCARC) Reduce Fatal Accident Rate . . . Strategic Plan to Improve Safety Improve Safety Worldwide . . . Both government and industry had been long been searching for a reliable way to choose wisely among the many deserving suggested actions to improve various aspects of aviation, thereby reducing risk. The Gore Commission’s challenge to reduce accidents by 80% over the next ten year provided the impetus for even closer collaboration to develop a data-driven process to help us focus our scarce resources on those initiatives with the most potential to reduce the risk of accidents. We were all convinced that the rigor of analyzing reliable data was the surest path to continually improving our performance. 2

4 Vision - Mission - Goals
Key aviation stakeholders acting cooperatively to lead the world-wide aviation community to the highest levels of global commercial aviation safety by focusing on the right things. Mission Enable a continuous improvement framework built on monitoring the effectiveness of implemented actions and modifying actions to achieve the goal. Goal Reduce the US commercial aviation fatal accident rate 80% by 2007. And Maintain a continuous reduction in fatality risk in US and International commercial aviation beyond 2007. The vision ,mission, and goals of CAST are as shown on this slide.

5 CAST brings key stakeholders to cooperatively develop & implement a prioritized safety agenda
Industry Government AIA Airbus ALPA APA ATA IFALPA NACA Boeing GE* RAA FSF DOD FAA Aircraft Certification Flight Standards System Safety Air Traffic Operations Research NASA ICAO** EASA (ECAST) TCC NATCA** NTSB** Commercial Aviation Safety Team (CAST) The strength of CAST lies in its extensive membership, its proactive commitment to safety and its ability to effect change. The CAST has proven effective because it is a voluntary association of key stakeholders in the operation of the commercial aviation system and the participants are safety leaders from those organizations able to commit and effect change These organizations have come together voluntarily to improve aviation safety: Aerospace Industries Association (AIA) Department of Defense (DOD) Air Line Pilots Association (ALPA) Federal Aviation Administration (FAA) Allied Pilots Association (APA) National Aeronautics and Space Administration (NASA) Air Transport Association (ATA) International Civil Aviation Organization (ICAO) National Air Carrier Association (NACA) European Aviation Safety Agency (EASA) General Electric (GE)-representing all engine manufacturers Transport Canada (TCC) Regional Airline Association (RAA) National Air Traffic Controllers Association (NATCA) (observer) Flight Safety Foundation (FSF) National Transportation Safety Board (NTSB) (observer) International Air Transport Association (IATA) (observer) Association of Asia Pacific Airlines (AAPA) (observer) Air Transport Association of Canada (ATAC) (observer) Association of Professional Flight Attendants (APFA) (observer) IATA** AAPA** ATAC** APFA** ACI-NA** * Representing P&W and RR ** Observer

6 Joint Implementation Measurement Data Analysis Team (JIMDAT)
Commercial Aviation Safety Team (CAST) CAST Joint Safety Analysis Teams (JSAT) Data analyses Joint Safety Implementation Teams (JSIT) Safety enhancement development Straightforward & rigorous process JSAT – Analyze data Identify problems or precursors Propose interventions against those problems (can be out of the box proposals) JSIT – Develop candidate safety enhancements Assess feasibility of interventions Group promising interventions into package of enhancements Develop Detailed Implementation Plans (DIPs) JIMDAT – Prioritization/Evaluation of Effectiveness Determine overall effectiveness of proposals some much more effective than others Identify synergies Recognize resource requirements Develop into integrated, prioritized package of enhancements to the aviation system for CAST review JIMDAT/JSIT interaction may be iterative to maximize effectiveness of the detailed implementation plans Master safety plan Enhancement effectiveness Future areas of study Joint Implementation Measurement Data Analysis Team (JIMDAT)

7 Safety Plan Development
Accident JSATS Case studies Accident JSITS Case studies Safety Enhancements Recommended Plan Revision Emerging Risk Master Contributing Factors Incident Analysis Process Develop Enhancements & Metrics JIMDAT Review CAST Plan Changing Risk Metrics Performance To Plan Review Non- Performance Information Aviation System Changes Identify Hazards Identify Factors We have completed the historical study of CFIT, Loss of Control, Approach and Landing, Runway Incursion and Turbulence accidents and hull losses which have occurred in U.S. FAR 121 operations over the time frame of 1987 to Additionally we have completed an assessment of accidents and hull losses world wide over that same timeframe. The yellow boxes, ‘Accident JSAT’s’, ‘Accident JSIT’s’, etc. depict this historical study of accidents from which CAST has identified safety solutions to proactively apply and prevent/mitigate recurrence. CAST is transitioning to an incident analysis process that will allow us to become more proactive in accident prevention by identifying changing and emerging risks. This is shown by the green boxes, ‘Incident Analysis Process’, ‘Emerging Risk’, ‘Changing Risk’, etc. Safety enhancements from this activity will be rolled into the CAST plan, related metrics will be developed and any newly identified contributing factors will be added to the Master Contributing Factor list. Also to reach further into the future (orange), CAST will examine and identify hazards that may result from ‘Aviation System Changes’ and ‘Demographic Changes.” Much of this work has been done by CAST’s sister organization, the European led Future Aviation Safety Team (FAST), which is developing future hazards based on their study of future areas of changes. CAST will incorporate the results from the FAST analysis into the CAST plan; safety enhancements and related metrics will be developed and the newly identified contributing factors will be added to the Master Contributing Factor list. Yes Present In Master Factors Develop Contributing Factors (new or emerging FAST Hazards No Demographic Changes Identify Hazards Identify Factors CAST-051R

8

9 Robust CAST Methodology
Detailed event sequence - problem identification from worldwide accidents and incidents Broad-based teams (45-50 specialists /team) Over 450 problem statements (contributing factors) Over 900 interventions proposed Analyzed for effectiveness and synergy Extremely robust and disciplined process Amazing array of talent brought to bear to develop CAST plan Evidence of commitment of participants to the CAST process

10 CAST Process Led to Integrated Strategic Safety Plan
Part 121 or equivalent passenger and cargo operations studied Current CAST plan: 72 Prioritized Safety Enhancements 50 Complete and 22 underway Projected 74% fatality risk reduction by 2020 Industry and Government implementing plan Acknowledge that other enhancements ongoing bring number to over 80%

11 Resource Cost Vs. Risk Reduction
100% 10000 Risk Reduction APPROVED PLAN $ 9000 Total Cost in $ (Millions) 8000 75% 7000 6000 Risk Eliminated by Safety Enhancements $ 50% Resource Cost ($ Millions) 5000 2007 2020 4000 3000 25% 2000 A graphical representation of resource application versus risk reduction, which also depicts the CAST JIMDAT prioritized selection criteria for the draft strategic plan. In the example plot, it can be seen how the CAST plan items for 2007 and 2020 were selected using benefits versus resources and the rationale for not selecting all the solutions. As you can see there was a definite knee in the resource cost curve. The selected CAST plan provided an estimated 74% risk reduction by 2020 at a cost of about 740 million dollars. To implement all of the CAST possible projects would increase the risk reduction by about 5 percent but require more than 5 billion dollars to implement. $ $ 1000 $ 0% Completed Completed + Plan (2007 Implementation Level) Completed + Plan (2020 Implementation Level) All JSIT Proposed Enhancements (2020 Implementation Level)

12 Cost Savings Dollars/Flt. Cycle
Part 121 Aviation Industry Cost Due to Fatal/Hull Loss Accidents 100 Historical cost of accidents per flight cycle 80 74% Risk reduction Savings ~ $74/Flight Cycle Or ~ $814 Million Dollars/Year Dollars/Flt. Cycle 60 40 When we break down costs of our current accident rate, accidents cost us approximately $100 for every flight. By implementing the 72 carefully selected, data-driven safety enhancements, we will have reduced these costs by $74 per flight. This adds up to a savings about $814 Million EVERY YEAR into the future. 20 Cost of accident fatalities following implementation of the CAST 2020 levels 2007 2020

13 CAST Safety Plan 51 Completed Safety Enhancements Safety Culture
Maintenance Procedures Flight Crew Training Air Traffic Controller Training Uncontained Engine Failures Terrain avoidance warning system (TAWS) Standard Operating Procedures Precision Approaches Minimum Safe Altitude Warning (MSAW) Systems Proactive Safety Programs (FOQA + ASAP) Although time does not permit my discussing each of the following 50 completed safety enhancements, this list illustrates the areas, and types of actions, that were covered in the enhancements completed to date. This list clearly indicates how a range of training, operational and design solutions were utilized. CFIT-PAI-Vertical Angles - Increases the use of Precision approach through addition of vertical angles on approach plates to achieve constant angle descent.

14 CAST Safety Plan (cont.)
21 Committed Safety Enhancements Policies and Procedures Aircraft Design Flight Crew Training (additional aspects) Runway Incursion Prevention Precision Approaches (additional projects) Icing (additional turboprop projects) Midair Maintenance Runway Safety Safety culture, policies and procedures Study of midair, maintenance, cargo and additional icing related accidents and serious incidents resulted in a number of new Safety Enhancements. The new Safety Enhancements are currently being implemented.

15 Fatal Accident Rate and Full Airplane Loss Equivalents
Rate for Part 121 Operations (5 year moving average) 9.0 8.0 5 year moving avg of fatal accidents per 10 million departures 7.0 6.0 5.0 Fatality Accidents or Full Loss Accident Equivalents per 10 Million Departures 4.0 3.0 82% Fatality Accident Rate Reduction Since we are past the original period, the question you might ask is – Did CAST reach its 2007 goal?? This chart illustrates the five year rolling average of the fatal accident rate in U. S. Part 121 operations (red line) and the fatality risk (blue line). This clearly shows that the CAST activity, coupled with other ongoing safety activities by government and industry, have resulted in an 82 percent reduction in the fatal accident rate in the ten years ending in 2007. 2.0 1.0 0.0 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

16 International Perspective CAST Safety Enhancements
Western-built transport hull loss accidents, by airline domicile, 2001 through 2007 COSCAP CIS C.I.S.1 Europe 0.7 ESSI 35 SEs JAA 11 SEs EASA 62% reduction COSCAP NA, SA, SEA 40 SEs in work 60% reduction United States and Canada 0.4 China 0.3 CAST 70 SEs 48 complete 22 in work 74% reduction Middle East 2.3 Asia 2.1 (Excluding China) COSCAP BAG Africa 10.0 COSCAP Gulf Latin America and Caribbean 2.1 Oceania 0.0 ASET RASG-PA World 1.1 Accidents per million departures 1 Insufficient fleet experience to generate reliable rate.

17 Summary History shows focused action and introduction of new capabilities have led to accident risk reductions Joint industry and government teams working together to a common goal can further enhance the safety of our very safe aviation system Full implementation will require a coordinated effort between industry and government CAST is moving forward to meet the challenge ASIAS (Aviation Safety Information Analysis and Sharing) initiative is under way help aviation safety teams (CAST, IHST, GAJSC) meet the challenge

18 What is ASIAS…. A collaborative Government-Industry initiative on data sharing & analysis to proactively discover safety concerns before accidents or incidents occur, leading to timely mitigation and prevention 18

19 ASIAS AIRLINE PARTNERS
AIRTRAN AIRWAYS ALASKA AIRLINES AMERICAN AIRLINES AMERICAN EAGLE ATLANTIC SOUTHEAST AIRLINES CHAUTAUQUA AIRLINES COMAIR COMMUTAIR COMPASS AIRLINES CONTINENTAL AIRLINES EXPRESSJET FRONTIER AIRLINES GULFSTREAM INTERNATIONAL AIRLINES JETBLUE AIRWAYS DELTA-NORTHWEST AIRLINES REPUBLIC AIRLINES SHUTTLE AMERICA SKYWEST AIRLINES SOUTHWEST AIRLINES SUN COUNTRY AIRLINES UNITED AIRLINES UNITED PARCEL SERVICE US AIRWAYS

20 ASIAS Participants at Major US Airports
20 20

21 ASIAS ASAP Records in Production
Record counts as of 14 Sept 2009 21 21

22 ASIAS FOQA Records in Production
As of Sept 2009, over 5 Million FFD Records available to ASIAS 22 22

23 Data Sources Supporting ASIAS InfoSharing and Analysis
De-Identified FOQA Data De-Identified ASAP Data Traffic Management Reroutes and Delays Airport Configuration and Operations Sector and Route Structure Procedures Surveillance Data for En Route, Terminal and Airport ATC Information Flight Operations Maintenance Dispatch ATSAP Safety Reports Aviation Safety Reporting System Runway Incursion Surface Incident Operational Error / Operational Deviation Pilot Deviation Vehicle or Pedestrian Deviation National Transportation Safety Board Accident/Incident Data System Service Difficulty Reports Bureau of Transportation Statistics Weather / Winds Manufacturer Data Avionics Data Worldwide Accident Data Other Information 23 23

24 ASIAS is Governed by Formal Principles
Data used solely for advancement of safety Non-punitive reporting Airline data is de-identified Analyses approved by an ASIAS Executive Board 24 24

25 ASIAS Studies Completed or Underway
Directed Studies TCAS Resolution Advisories Runway Safety Terrain Awareness Warning Unstable Approaches Analysis In Progress Analysis Completed Airline Benchmarks Terrain Awareness Warning Airline Stability Metric TCAS Resolution Advisories CAST Metrics: Known Risk Monitoring Safety Enhancement Assessment Approach and Landing Controlled Flight Into Terrain Loss of Control Mid-Air Collision Icing Cargo Maintenance 25 25

26 Analytical Products Terrain Awareness Warning System (TAWS)
Traffic Collision Advisory System (TCAS) Wrong Runway Departures (WR) Metrics

27 Data Fusion Provides Valuable Insights
Traffic Tracks Source: FAA National Offload Program Safety Event Focus Source: Digital Flight Data, Safety Reports Minimum Vectoring Altitudes Source: Air Traffic Control Airport & Airspace Procedures Source: Air Traffic Control Weather Source: FAA & NOAA Terrain Source: National Elevation Data

28 Bay Area Traffic Flows – SE Plan

29 Safety Enhancement 184 – Minimum Vectoring Altitude Reevaluation
Reevaluate minimum vectoring altitudes (MVAs) at prioritized sites identified in the CAST TAWS study to reduce the number of terrain awareness warning system (TAWS) alerts. ASIAS has developed a tool to identify MVA’s that should be reviewed. The CAST mitigation team is coordinating the evaluation of this tool within the FAA. 29 29

30

31 Safety Enhancement 185 – TAWS and RNAV Visual or Other Procedures
This enhancement provides better separation from terrain by providing RNAV Visual or other procedures that mitigate known TAWS and terrain issues. jetBlue has agreed to be the lead carrier in developing a RNAV visual flight procedure for OAK Multiple airports identified in the TAWS study have implemented, or are developing, visual and/or instrument flight procedures ABQ: RWY 26 RVFP in place RWY 26 RNP waiting for completion of sight survey LAS: RWY 19 RVFP (USAirways & NetJets) SAN: RWY 27 RNP SAR having issues with survey & environmental Guam: Continental talking about RVFP (no formal application) 31 31

32 OAK Traffic – RNAV Capable
Aircraft RNAV equipped per AC A DME/DME/IRU and/or GPS sensor input NOTE: Explain adequate equipage levels and weather suitability for procedure during this slide. 32

33 Proposed: RNAV VISUAL Approach to OAK RWY 11
33

34 Safety Enhancement 120 – TAWS Improved Functionality
Current production models, new type design airplanes, and existing aircraft, where appropriate, will include GPS equipment to allow incorporation of certain TAWS enhancements. CAST is developing a cost benefit analysis to help promote this initiative 34 34

35 EGPWS Mode-2 Pull Ups – Pre 218
35

36 EGPWS Mode-2 Pull Ups w/218 or Higher (about 20X Less Frequent)
36

37 Analytical Products Terrain Awareness Warning System (TAWS)
Traffic Collision Advisory System (TCAS) Wrong Runway Departures (WR) Metrics

38 Data Sources and Analysis
NOP Run simulations that will encompass all carriers and traffic to better estimate RA rates and degree of risk. Perform domestic / foreign comparisons. (Provides intruder information.) (National Offload Program radar surveillance data) FOQA Assess NAS-wide and airport-specific TCAS RA landscape. Provide Trend Information. Investigate pilot response to advisories. (TCAS Operational Performance Assessment) TOPA Mode-S downlink provides information similar to FOQA, and is enhanced with surveillance radar data. Four primary data sources All are valuable, but all have limitations… NOP -- Surprisingly difficult to get high fidelity data. Only got one month of data for each target airport. Data must be run through simulator to estimate RAs FOQA – Highly detailed for single aircraft, but de-identification makes it impossible to see the context and see the intrude -- Includes only the five ASIAS airlines with data in the FOQA system TOPA – A new data source available at only a few airports. Big advantage is that it has radar (for intruder info) and “hears” actual Ras Text Reports – Difficult to use quantitatively because of self-report bias. Text Reports Help get the story behind the numbers at selected airports. (ASAP, ASRS, OE) 38 38

39 ASIAS Directed Study on TCAS RAs: Location from FOQA Data
Over a three-year period, Southern California, Northern California, and Denver accumulated the greatest concentrations of TCAS RA events. * 39 39

40 Common Themes Observed at Study Airports: VFR and IFR Traffic Interaction
GA aircraft under tower control interacting with structured* IFR traffic at a nearby airport Example: BUR arrival and departure interaction with VNY “Loitering” VFR traffic interacting with structured IFR traffic to/from nearby airport Example: Philadelphia downtown VFR traffic interaction with PHL arrivals Types of traffic observed in data: radio / TV traffic reporting airplanes and helicopters, aerial photography, law enforcement Structured VFR flow near structured IFR traffic Example: VFR helicopter route crossing under EWR arrivals Example: GA traffic crossing under JFK arrivals *Arrival /departure route or an approach Looking across the airports, a number of common themes emerged. This slide shows the VFR vs IFR themes (next slide is IFR vs. IFR) TCAS will often sound an RA at 600’ vertical separation, but 500’ is “legal” but you’ll still get an RA 500-ft separation is maintained in most cases, but it is not sufficient to prevent TCAS RAs 40 40

41 Common Themes Observed at Study Airports: IFR Traffic Interacting with Other IFR Traffic
Visual approaches to parallel runways Close spacing Example: SFO 28R/L with 750’ centerline spacing High altitude with increased TCAS sensitivity Example: DEN 16R/L with 2500’ centerline spacing Arrival / departure interaction triggered by closure rate Example: TEB west departures with EWR 22 arrivals Example: DEN west departures with arrivals on both north and south flows This slide shows the IFR vs IFR themes seen across airports Parallel runways at SFO and DEN were big -- Keep in mind some airlines at SFO shift to TA-only TCAS sensitivity decreases as altitude decreases ( to prevent unwanted RAs) Baro Pressure is used to determine altitude until radar altimeter kicks in. DEN parallels may be experiencing “over sensitivity” before radar altimeters kick in. Because TCAS can’t take INTENT into consideration, a departure climbing to and an arrival descending to could set off an RA These procedures are in accordance with , but they do not provide sufficient separation to prevent TCAS RAs 41 41

42 Analytical Products Terrain Awareness Warning System (TAWS)
Traffic Collision Advisory System (TCAS) Metrics Wrong Runway Departures (WR)

43 Mapping Problem Statements to FOQA Metrics
The mapping of Problem Statements to actual metrics that can be measured in FOQA data is a complex task. It is not a one-to-one relationship, and in many cases problem statements map to multiple metrics and individual metrics map to multiple problem statements. 43 43

44 Example Metrics for Approach and Landing Accident Risks
Metric Category Level Safety Enhancement Level Problem Statement Level Metric Level 1 Accident Spotlight Air Carrier: Cayman Airways Aircraft Type: B Location: Grand Cayman, Cayman Is. Synopsis: Night; VMC; Experienced crew; Visual approach; Light rain; Tailwind; 7,021 ft. runway; Thrust reversers inoperative; Speedbrakes not armed or deployed; Fast approach; 3.5° descent angle 1 Unstable approaches are measured using the Height Above Touchdown (HAT) method, which captures the HAT on the first approach when the aircraft is 100% stable The same approach was followed to analyze each Problem Statement. This is an excerpt from the MindMap with an Accident Spotlight. The accidents provide specific crash-related metrics that can then be analyzed using FOQA events and measurements. This slide connects Safety Enhancement #23 to Problem Statements 10, 17, and 48 to Did not Fly Stable Approach and Did not Manually Deploy Speedbrakes. The Accident Spotlight summarizes the description that would be found on the MindMap by clicking the “paperclips” marked with red 1s. Stable on Glideslope Stable on Localizer Stable [CAS-Vapp] Stable [CAS-Vref] Stable Ground Speed Stable Rate of Descent (ROD) Stable Thrust Landing Flaps Set Gear Down Stable in Pitch Stable in Roll Stable in Yaw Ground Spoilers Armed Speed Brakes Retracted 44 44

45 Initial Results: Unstable Approach Metric
83.1% 15.8% 1.1% This slide, reading from right to left, shows that 83.1% of flights were stable by 1,000 feet HAT % of flights became stable between 1,000 feet and 500 feet above touchdown. 1.1% of flights remained unstable below 500 feet above touchdown. The histogram segments, the colors, show which stable approach event became stable last. The tallest bar shows that gear, flaps, thrust, airspeed, on-localizer and on-glideslope were most often last above 1,000 feet. Between 1000 and 500 feet on descent, speed and glideslope are not prominent. The most frequently occurring events in the 1.1% column includes: Low Thrust, Glideslope, Localizer, and Flaps events. And the most frequently occurring of those events is Low Thrust (approximately 0.5% of flights) 45 45

46 Initial Results: Unstable Approach Metric
The previous slide trends Unstable Approach (HAT Method). This one trends Unstable Approach (HAT Method) AND Go-Around. 46 46

47 Analytical Products Terrain Awareness Warning System (TAWS)
Traffic Collision Advisory System (TCAS) Metrics Wrong Runway Departures (WR)

48 Lexington Blue Grass (LEX) 2006
Aircraft was cleared for departure on Runway 22 but departed on Runway 26 Comair flight 5191 crashed approximately ½ mile from the end of runway 26 Similar non-fatal events have occurred prior to this event Cleared for 22 but lined up on 26 (1993) Poor visual cues and lighting also cited in other taxing related events by air crews Similar non-fatal events have occurred after this event

49 Findings – Part 121 Operations
Wrong Runway Departures By Data Source NOTE: ASRS database with certain exceptions captures ~ 18% of reports received by NASA on monthly basis 10 20 30 40 50 60 70 80 90 ASRS PDS NTSB (I) AIDS (I) PTRS NTSB (A) 84 32 1

50 Findings – Part 121 Operations
Air Carrier Reported Wrong Runway Events ( ) Cleveland Houston Salt Lake City Miami Chicago Lexington NOTE: ASRS database captures only 18% of all reports received by NASA on monthly basis 5 10 15 20 25 30

51 Part 121 Contributing Factors

52 JIMDAT Mitigation Assessment
Risk Eliminated 52 52

53 Findings – Part 135 Operations
Wrong Runway Events ( ) Cleveland Houston Lexington 5 10 15 20 25 30 Houston PART 121 Cleveland Los Angeles Richmond Syracuse Anchorage NOTE: ASRS database with certain exceptions captures ~ 18% of reports received by NASA on monthly basis

54 Part 139 (Class 1) Airport Review
355 Airports

55 International Collaboration
Sharing of CAST safety products worldwide – COSCAP, GASR, etc CAST ICAO Common Taxonomy ECCAIRs – common taxonomy, developing data sharing proposals TCAS RA mitigation study – including Eurocontrol Desire increased connectivity with ICAO ISDCAS (Integrated Safety Data Collection and Analysis System) IATA GSIC – Evaluating sharing top level safety hazards Partnered with E-CAST

56 QUESTIONS? http://www.cast-safety.org/
To meet the recommendations of the recommendations of the White House Commission on Aviation Safety and the National Civil Aviation Review Commission the Existing Industry Safety Strategy Team (ISST) and the Federal Aviation Administration Safety Action Team (SAT) were combined, and augmented, to create the Commercial Aviation safety Team (CAST) in 1998.


Download ppt "Improving Safety through Collaborative Safety Initiatives"

Similar presentations


Ads by Google