Faculty Advisor Dr. Lance Sherry Sponsor Integrity Applications Incorporated Design of a System for Aircraft Fuselage Inspection Rui Filipe Fernandes Kevin Keller Jeffrey Robbins Reduce Inspection Time Improve Crack Detection Reduce Maintenance Cost jchadwickco.com Before Crack gasolinealleyantiques.com aerospacetestinginternational.com After Computer Aided Detection Manual Inspection Automated Inspection Fatigue Damage mechanicsupport.blogspot.com
57
Embed
Design of a System for Aircraft Fuselage Inspection...Widespread Fatigue Damage Design of a System for Aircraft Fuselage Inspection WFD Leading to Aircraft Retirement . Design of a
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Faculty Advisor Dr. Lance Sherry
Sponsor Integrity Applications Incorporated
Design of a System for Aircraft Fuselage Inspection Rui Filipe Fernandes Kevin Keller Jeffrey Robbins
Reduce Inspection Time Improve Crack Detection
Reduce Maintenance Cost
jchadwickco.com
Before
Crack
gasolinealleyantiques.com
aerospacetestinginternational.com
After
Computer Aided Detection
Manual Inspection Automated Inspection
Fatigue Damage
mechanicsupport.blogspot.com
Design of a System for Aircraft Fuselage Inspection 2
Agenda
Context • Aging Aircraft & Maintenance • Current Fuselage Inspection Process • Stakeholder Analysis • Problem and Need
Concept Of Operations Method of Analysis
Project Plan
3
Context: Aging Aircraft & Maintenance Aircraft Age Statistics
Min: 5.1 years Mean: 10.6 years Max: 24.9 years airsafe.com
Design of a System for Aircraft Fuselage Inspection
Average Aircraft Age Continues to Increase
4
iata.org
Context: Aging Aircraft & Maintenance Increasing Average Age
Design of a System for Aircraft Fuselage Inspection
Rank Carrier Average Age
4 US Airways 14.7
5 Southwest 14.6
6 United 13.7
International Air Transport Association Bloomberg
Domestic Carriers among the oldest fleets
bloomberg.com
5
iata.org travelpulse.com
Widespread fatigue damage (WFD) is age-related structural fatigue cracking
Design of a System for Aircraft Fuselage Inspection
Maintenance Intervals, A Delicate Balance of Risk and Cost
Context: Aging Aircraft & Maintenance Median Crack Growth Curve
11 Design of a System for Aircraft Fuselage Inspection
Minimize Number of Cracks Occurring Before Inspection
Earliest Expected Cracking
Latest Expected Cracking
Time of Inspection
Critical Crack Length
Cra
ck L
engt
h
Time
Median crack growth curve
Yang JN, Manning SD (1990) Stochastic Crack Growth Analysis Methodologies For Metallic Structures
Context: Aging Aircraft & Maintenance Distribution of Time to Critical Crack Length
12 Design of a System for Aircraft Fuselage Inspection
Time of Inspection
Critical Crack Length
Probability of cracks occurring BEFORE scheduled Maintenance
Minimize Probability of Cracks Occurring Before Inspection
Cra
ck L
engt
h
Time
Distribution of Time to Critical Crack Length
Yang JN, Manning SD (1990) Stochastic Crack Growth Analysis Methodologies For Metallic Structures
Context: Aging Aircraft & Maintenance Distribution of the Crack Length
13 Design of a System for Aircraft Fuselage Inspection
Probability of crack growth beyond critical length
Critical Crack Length
Time of Inspection
Minimize Crack Growth Beyond Critical Length
Distribution of the crack length
Cra
ck L
engt
h
Time
Yang JN, Manning SD (1990) Stochastic Crack Growth Analysis Methodologies For Metallic Structures
Context: Aging Aircraft & Maintenance Stochastic Crack Growth Model
14 Design of a System for Aircraft Fuselage Inspection
Probability of crack growth beyond critical length
Critical Crack Length
Time of Inspection
Probability of cracks occurring BEFORE scheduled Maintenance
Earliest Expected Cracking
Latest Expected Cracking
Early Crack Detection Can Minimize Corrective Maintenance
Distribution of Time to Critical Crack Length
Distribution of the crack length
Median crack growth curve
Cra
ck L
engt
h
Time
Yang JN, Manning SD (1990) Stochastic Crack Growth Analysis Methodologies For Metallic Structures
Context: Aging Aircraft & Maintenance Stochastic Crack Growth Model
15 Design of a System for Aircraft Fuselage Inspection
The inspection schedule is chosen such that the probability of crack to grow beyond the critical crack size is less than 1 in 10,000,000
Taghipour, S., Banjevic, D., Jardine, A. K. S., “Periodic inspection optimization model for a complex repairable system”, Reliability Engineering and System Safety, Vol 95, 2010, Pg 944-952
Yang JN, Manning SD (1990) Stochastic Crack Growth Analysis Methodologies For Metallic Structures
16
When it finds an unsafe condition exists in the product and the condition is likely
to exist or develop in other products of the same type design
Airworthiness Directives are legally enforceable regulations issued by the Federal Aviation Administration (FAA) in accordance with 14 CFR part 39 to correct an unsafe condition in a product
faa.gov
Design of a System for Aircraft Fuselage Inspection
Corrective Maintenance is Disruptive to Airlines and Results in Unplanned Revenue Loss
Context: Aging Aircraft & Maintenance Title 14 of the Code of Federal Regulations (CFR)
17
faa.gov
Design of a System for Aircraft Fuselage Inspection
Inspection Process Governed by Title 14 (CFR)
Changes in maintenance procedure is regulated by the FAA
18
Context: Current Fuselage Inspection Process Visual Inspection Process
• Job Cards Used For Every Component
• Many Human Factors/Prone to Errors
• 41.8% detected • 14.1% type 1 error (Misdiagnosed) • 43.7% type 2 error (Missed Detection)
• Non-Destructive Inspection (NDI)
methods used to assess marked regions
Design of a System for Aircraft Fuselage Inspection
Inspection Process Begins with Visual Inspection
VISUAL INSPECTION RESEARCH PROJECT REPORT ON BENCHMARK INSPECTIONS FAA Aging Aircraft NDI Validation Center
19
Context: Current Fuselage Inspection Process Representative Regions of Aircraft
FAA Aging Aircraft NDI Validation Center Report
JC 501 Midsection Floor
JC 502 Main Landing Gear Support
JC 503 Midsection Crown (Internal)
JC 504 Galley Doors (Internal)
JC 505 Rear Bilge (External)
JC 506 Left Forward Upper Lobe
JC 507 Left Forward Cargo Compartment
JC 508/509 Upper and Lower Rear Bulkhead Y-Ring
JC 510 Nose Wheel Well Forward Bulkhead
JC 511 Lap-Splice Panels
Design of a System for Aircraft Fuselage Inspection
ntl.bts.gov
Representative Regions Require Different Inspection Techniques
20
Context: Current Fuselage Inspection Process Current Visual Inspection Process
Design of a System for Aircraft Fuselage Inspection
Inspection Process Modeled In Simulation
95704520
6
5
4
3
2
1
0
Shape 5.008
Scale 9.652
N 12
Inspection Time (Minutes)
Fre
qu
en
cy
Gamma
Inspection Time of Lap-Splice Panels (Minutes)
VISUAL INSPECTION RESEARCH PROJECT REPORT ON BENCHMARK INSPECTIONS FAA Aging Aircraft NDI Validation Center
21
Context: Stakeholder Analysis Interactions, Tensions and Gap
Design of a System for Aircraft Fuselage Inspection
Context: Problem and Need
22 Design of a System for Aircraft Fuselage Inspection
Issues Consequences
Heavy D Check Inspection Process Requires up to 2 months to Complete
Aircraft maintenance/repair 12-15% of total airline annual expenditures
In 2013, 3.5 million flight cycles logged over 2,660 aircraft
Average $2,652 per flight cycle Amounts to $9.4 billion total
43.7% Type 2 Error (Missed Detection) 11 Airworthiness Directives Issued to Address Fuselage Cracking
Solutions Benefits
Reduce Time Required for Inspections Decreased Inspection Costs
Early Detection of Structural Fatigue Improved Scheduling of Preventive Maintenance / Minimize Corrective Maintenance Required
Reduce Human Error Improved Crack Detection Capabilities
Problem
Need
Current Inspection Process
Improved Inspection Process
Time
Cost
Quality
Win-Win: New Technology Introduced to Inspection Process
23
Agenda
Context Concept of Operations
• Operational Scenario • Design Alternatives • System and Design Requirements • Automated Inspection System IDEF.0
Method of Analysis Project Plan
Design of a System for Aircraft Fuselage Inspection
24
Concept of Operations: Operational Scenario Levels of Human Involvement
Design of a System for Aircraft Fuselage Inspection
Inspection Method
ConOps Introduces New Technology to Inspection Process
1 – Manual 2 - Enhanced
ntl.bts.gov aviationpros.com aviationpros.com
3 - Autonomous
25
Concept of Operations: Operational Scenario Non-Contact Delivery Method
Design of a System for Aircraft Fuselage Inspection
Potential Implementation of Synthetic Aperture Imaging Technology
Track
Synthetic Aperture Imaging Device
26
Concept of Operations: Design Alternatives Exterior vs. Interior Surfaces
Exterior Surfaces Interior surfaces
Human Visual Human Visual
Human Remote Visual
Human Enhanced Visual Human Enhanced Visual
Robotic Crawler*
Non-Contact Automated Scan*
* Utilizes Image Processing Software
Design of a System for Aircraft Fuselage Inspection
Limitations of Delivery Method Based on Region of Aircraft
27
Inspection Method
Time Cost Quality
Visual Visual Inspection time Documentation time
Hourly wage of inspectors Training Cost Cost of Human Errors
Limited by human eyesight Prone to human error Human decision making only
Enhanced Visual
Increased Inspection Time Imaging Time Evaluation Time Documentation Time
Hourly wage of inspectors Training/certification Maintenance Cost Cost of Human Errors
Improved by computer
aided decision making Interpretation/ Evaluation of data prone to human errors
Automated Faster Inspection Time
Imaging Time
Software Processing
Time
Acquisition/Development Cost Installation Cost Training Cost Maintenance Cost
Software for image
processing reduces
errors and eliminates
dependence on human
decision making
PRO CON
Concept of Operations: Design Alternatives Benefits by Category
Design of a System for Aircraft Fuselage Inspection
28
Context: System Requirements Mission & Functional Requirements
Mission and Functional Requirements
M.1 The system shall reduce the airframe maintenance cost per flight hour of an aircraft by 5%
F.1. The system shall cost no more than $X to operate annually
F.2. The system shall accrue no more than $X in Type 1 errors annually
F.3. The system shall require an initial investment of no more than $X
F.4. The system shall process captured images at a rate of X m2 per Y seconds
M.2 The system shall detect cracks in the airframe of aircraft both visible, and not visible, by a human
inspector
F.1. The system shall detect cracks with a volume exceeding X mm3
F.2. The system shall have a Type 2 error rate of no more than X%
F.3. The system shall distinguish between cracks and pre-built parts of the aircraft
F.4. The system shall capture an image of the airframe of the aircraft of dimensions X meters by Y meters
without repositioning
M.3 The system shall reduce the variance of the airframe inspection process by X labor-hours
F.1. The system shall maintain the upper bound of a complete visual inspection at no more than X labor-
hours
F.2. The system shall reduce the variance of the visual inspection process by X labor-hours
M.4 The system shall allow aircraft to meet Federal Aviation Administration airworthiness standards
Context: System Requirements
Design of a System for Aircraft Fuselage Inspection
29
Non-Functional Requirements
Maintainability
1. The system shall produce traceable error codes upon malfunction.
2. The system shall allow the replacement of individual parts.
Reliability
1. The system shall experience no more than X system failures per month.
2. The system shall require no more than X hours of preventative maintenance per month.
Usability
1. The system shall require no more than 40 hours of training for technician certification.
Context: Non-Functional Requirements
Design of a System for Aircraft Fuselage Inspection
30
Concept of Operations: Design Requirements
Design of a System for Aircraft Fuselage Inspection
Design Requirements
Enhanced Visual (Handheld)
D.1 The system shall weigh no more than X lbs.
D.2 The system shall accurately scan from a distance of up to X m.
Robotic Automated Inspection System
D.1 The system shall inspect at a rate of X cm3/s.
D.2 The system shall support autonomous function.
D.3 The system shall accept initial input from an operator.
D.4 The system shall utilize integrated software.
D.5 The system shall store the location of airframe problem areas.
31
Concept of Operations: Automated Inspection System IDEF.0
Design of a System for Aircraft Fuselage Inspection
32
Agenda
Context Operational Concept/Approach Method of Analysis
• Stochastic Simulation • Model Boundaries & Simulation Inputs/Outputs • Simulation Requirements • Simulation of Visual Inspection By Airframe Region • Case Study Variables & Assumptions • Validation
• Design of Experiments
Project Plan
Design of a System for Aircraft Fuselage Inspection
33
Method of Analysis: Stochastic Simulation Model Boundaries and Simulation Inputs/Outputs
Design of a System for Aircraft Fuselage Inspection
Inputs Outputs
• What design alternatives are utilized • Where design alternative are utilized
• Overall time for inspection • Time per section • Cracks detected per section • Type 1 errors per section • Type 2 errors per section
Aircraft Maintenance
Simulation
Uninspected aircraft
Inspected aircraft
• Time per inspection • Inspection & Section
• Cost per inspection • Labor hours • Implementing alt.
• Quality per inspection • Type 1 & 2 errors
Manual • Human • Handheld
Automated • Visual or thermal • Track or crawler
34
Method of Analysis: Stochastic Simulation Simulation Requirements
Simulation Requirements
The simulation shall break down the aircraft into ten sections, each having its own queue
The simulation shall support multiple inspectors processing multiple sections
The simulation shall assign a set number of cracks to each section of the aircraft
The simulation shall terminate upon the inspection of all ten sections of the aircraft
The simulation shall collect statistics on total time required for inspection
The simulation shall collect statistics on total time to complete each section
The simulation shall collect statistics on cracks detected per section
The simulation shall collect statistics on crack type one errors
Mark a crack where one would not register with an NDT
The simulation shall collect statistics on crack type two errors
Fail to mark a crack that exists
Design of a System for Aircraft Fuselage Inspection
35
Method of Analysis: Stochastic Simulation Visual Inspection By Airframe Region
Design of a System for Aircraft Fuselage Inspection
Initialization
Process
Statistics
36
Method of Analysis: Stochastic Simulation Initialization
Design of a System for Aircraft Fuselage Inspection
Assignments
Manual / Automated (binary)
Process Restrictions (binary)
Process Distributions (minutes)
Crack Detection Rate (%)
Type 1 Error Rate (%)
Type 2 Error Rate (%)
37
Method of Analysis: Stochastic Simulation Process
Design of a System for Aircraft Fuselage Inspection
38
Method of Analysis: Stochastic Simulation Statistics
Design of a System for Aircraft Fuselage Inspection
39
Method of Analysis: Stochastic Simulation Distributions At a Glance
Design of a System for Aircraft Fuselage Inspection
VISUAL INSPECTION RESEARCH PROJECT REPORT ON BENCHMARK INSPECTIONS FAA Aging Aircraft NDI Validation Center
40
Method of Analysis Design of Experiments
Design of a System for Aircraft Fuselage Inspection
Run Internal /
External
Technology Delivery Method
Type One Error Rate
Type Two Error Rate
Inspection Time
Cost of Inspection
1 Internal Human -- % % Hours Dollars
External Human --
2 Internal Human --
External Thermographic Crawler
3 Internal Synthetic Aperture Handheld
External Thermographic Crawler
…
Inputs Outputs
41
Simulation Preliminary Results Sample Output (Time per Process)
Section Minutes Half-Width
Design of a System for Aircraft Fuselage Inspection
42
As-Is Simulation Preliminary Results Validation (Expected vs Simulation)
Section Actual (mins) Simulated (mins) Diff (mins) % err
1 122 116.47 -5.53 -4.53
2 28 27.83 -0.17 -0.61
3 75 75.38 0.38 0.51
4 68 67.71 -0.29 -0.43
5 37 36.1 -0.9 -2.43
6 104 105.64 1.64 1.58
7 95 100.23 5.23 5.51
8 35 34.68 -0.32 -0.91
9 16 15.2 -0.8 -5.00
10 48 49.56 1.56 3.25
Design of a System for Aircraft Fuselage Inspection
Actual Total (mins) Sim Total (mins) Diff (mins) % err
628 628.81 0.81 <0.1%
Sim Half Width (mins)
6.18
43
Agenda
Context Operational Concept/Approach Method of Analysis