|Unit Name||DATA ANALYSIS AND STATISTICS|
|Unit Duration||1 Term (online) or 1 Semester (on-campus)|
Graduate Diploma of Engineering (Safety, Risk and Reliability)
Duration: 1 year
Master of Engineering (Safety, Risk and Reliability)
Duration: 2 years
|Unit Creator / Reviewer||Keerthy Mysore|
Grad Dip total course credit points = 24
(3 credits x 8 (units))
Masters total course credit points = 48
(12 credits (Thesis) + 3 credits x 12 (units))
|Mode of Delivery||
Online or on-campus.
Combination of modes: Online synchronous lectures; asynchronous discussion groups, videos, remote and cloud-based labs (simulations); web and video conferencing tutorials. High emphasis on personal and group self-study.
Total student workload including “contact hours” = 10 hours per week:
Lecture – 1 hour
Tutorial Lecture - 1 hour
Practical / Lab - 1 hour (if applicable)
Personal Study recommended - 7 hours
Unit Description and General Aims
This unit focusses on Interpretation of Failure rates. Starting with a discussion on the differences between success and failure domains, the unit meanders through various topics such as sources of failure data, accuracy, confidence levels, Cumulative Poisson curves, Chi-square tests, and the bath-tub curve for constant failure rates and the Weibull distribution for variable failure rates. The unit concludes that the extremely wide variability of failure rates of identical components under identical conditions obviates the need for complex reliability prediction models.
On successful completion of this Unit, students are expected to be able to:
1. Compare and contrast failure data from various sources in terms of reliability, accuracy and confidence levels.
2. Apply correction factors to failure data to make it “fit for purpose”
3. Apply appropriate failure data to real-life situations
4. Employ appropriate statistical methods for data analysis and deductions
5. Interpret system failure rates and appreciate their limitations
6. Discuss Field data collection and Reliability Growth Modelling
(e.g. Assignment - 2000 word essay (specify topic) Examination (specify length and format))
(eg Week 5)
|Weighting (% of total unit marks)||Learning Outcomes Assessed|
Word length: n/a
Topic: Fundamental concepts of Failure and Reliability data and statistical methods overview
|Week 4||20%||1, 2|
Assessment 2 - mid-semester test
Type: Report (Midterm Project)
[This will include a progress report; literature review, hypothesis, and methodology / conclusions]
Word length: 1000
Topic: A comparative study on failure data from various sources e.g., field failures, manufacturer’s data and cycle testing, with an attempt to account for the differences noticed.
|Week 7||25%||1, 2, 3, 4|
Type: Report (Final Project)
[If a continuation of the midterm, this should complete the report by adding sections on: methodology, implementation / evaluation, verification / validation, conclusion / challenges and recommendations / future work. If this is a new report, all headings from the midterm and the final reports must be included.]
Word length: 2000
Topic: Report on statistical techniques used in Failure data analysis and the practical aspects of Reliability Growth Modelling.
|Week 12||35%||5, 6|
Students will be presented curves and plots and will be quizzed on interpretation.
Prescribed and Recommended readings
D. J. Smith, K. G. L. Simpson, Safety Critical Systems Handbook (4th Edition) - A Straightforward Guide to Functional Safety, IEC 61508 (2010 Edition), IEC 61511 (2016 Edition) and Related Guidance - Including Machinery and other Industrial Sectors
D. J. Smith, Reliability, Maintainability and Risk, 9th Edition. Elsevier, 2017 - ISBN: 978-0081020104
Number of peer-reviewed journals and websites (advised during lectures). Some examples are listed below.
- Reliability growth – Statistical test and estimation methods, Australian standard AS IEC 61164-2008.
- Equipment reliability – Reliability assessment methods, Australian standard AS IEC 62308– 2008.
- IDC /EIT notes and Reference texts as advised.
- Other material advised during the lectures
One topic is delivered per contact week, with the exception of part-time 24-week units, where one topic is delivered every two weeks.
Topic 1 and 2
1. Success and failure – domains and interpretation
2. Terms and jargon
3. Data accuracy, databases and confidence limits
Topic 3,4,5 and 6
Methods and mathematics
1. Statistical methods - Overview
2. Cumulative Poisson curves
3. Chi-square Distribution
4. Probability mathematics
Topic 7 and 8
Failure rates and methodologies
1. System failure rates
2. Methods - Overview
Topic 9 and 10
Failure and reliability modelling
1. Field failure data collection
2. Reliability Growth Modelling
Topic 11 and 12
Project and Unit Review
In the final weeks students will have an opportunity to review the contents covered so far. Opportunity will be provided for a review of student work and to clarify any outstanding issues. Instructors/facilitators may choose to cover a specialized topic if applicable to that cohort.
Project – due at end of term
Additional resources or files: N/A