Blog Archives

The Seven Reasons for Non-Linear Weibull Behavior

1. The Bath Tub Curve
A sample in question may be exhibiting one or more of the stages of the “bath tub curve” while it operates on test. This situation applies equally well to electronic or mechanical components or systems. A variety of failure modes are typically exhibited through the life history. The early life failure stage might exhibit a Weibull slope (Beta) of 0.6, followed by a mid-life stage with a slope of about 1.0. The wear phase has a Weibull slope typically greater than 1.5 and sometimes as high as 5.0. Some of the modes associated with early life failures, or infant mortality type failures, but other failure modes are usually associated with middle-life or end-of-life failures, the data can’t be linear. Further work and investigation is typically required to verify this situation, once a non-linear situation is recognized.
2. A Mixed Population
A sample in question may have been drawn from more than one sub-populations. These sub-populations often have distinct failure modes that are exhibited on Weibull graph as an “S-shaped curve” on the Weibull graph. Not all of the “S curves” may be visible due to sample size restrictions or even short test times. An example of the S curve follows.
3. Varying Environmental Conditions
This possibility may occur when test units or field systems operate in different environmental conditions. This is normally inadvertent or accidental and is often not discovered until data is plotted upon the Weibull graph and questions asked. The situation leads typically to a bimodal or multimodal result on the Weibull graph.
4. Mixed-Age Parts
Since most parts and systems do not carry a time clock, we cannot easily tell how old or how aged a part or system may be just by looking at it or putting it on test. Imagine for a moment a collection of aged (already have operated about 500 hours) mixed with new parts and all placed on life test or in operation. The test results would probably look a lot like a mixed population Weibull graph. The difference here is that the age difference is the main reason for differing sub-populations.
5. Three Parameter Weibull
Some parts (or systems) seem to have a natural bias concerning time-to failure. Examples include many material strength situations, car tires, or telephone poles. This situation leads to non-straight lines because of the natural offset present. Once this offset is recognized and corrected by a software program, the curves often straighten out into one smooth line.
6. Odd Distributions
Some distributions do not appear as straight lines on a Weibull graph. The most prominent example is the LogNormal distribution, which often appears as two straight lines which join at or very near 50% cumulative failures. Test this possibility by plotting data on Lognormal plot or another distribution.
7. Mixed Failure Modes
Very different failure modes, if operational during a test time or study period, may lead to unusual lines on the Weibull graph. Each line is usually associated with dominant failure mode. When modes are separated, as is customarily done, each failure mode usually appears as a straight line.

Published in Practical Weibull Analysis Techniques – Fifth Edition by James A. McLinn Published by  The Reliability Division of ASQ – January 2010 ISBN 0277-9633 (available as free download for ASQ Reliability Division Members)

“The Eight Reasons for Non-Linear Weibull Behavior”

 

Picture © B. Poncelet https://bennyponcelet.wordpress.com

Posted in General

TECH SPOT: SAMPLE CRE QUESTIONS (Part 2)

 

1. The predicted reliability is higher than the long term realized probability. Which of the following is the MOST likely cause of this difference ?

A. Deterioration of the manufacturing processes and procedures.

B. Lack of adequate employee training and process audits.

C. The accumulation of random process variations.

D. A poor initial estimation of reliability.

 

2. In a corrective action system, trend analysis can be defined as :

I. Long-term movement. III. A cyclical component of a time series.

II. The short term status of problems. IV.. Seasonal variations.

A. I only

B. II and III only

C. I, II and III only

D. I, II, III and IV

 

3. The MOST valid source of failure rate data is :

A. Test data obtained under very closely controlled conditions.

B. Environmental test data.

C. The manufacturing process.

D. Operational data.

 

4. A comprehensive failure analysis and corrective action feedback loop must determine :

I. What failedII. How it failedIII. Why it failed.

A. I only

B. I and II only

C. II and III only

D. I, II and III

 

5. What is this system’s reliability at 700 hours?

Where component failure data is :

– Failure rate of A = 0.0007 failures/hrReliability of B = 0.92

– MTTF of C = 1400 hours – Reliability of D = 0.85.

A 0.986

B. 0.998

C. 0.994

D. 0.952

 

6. Given mean-time-to-failure of 200 hours for each of two components, what is the probability of system failure if both components operate in parallel for one hour ?

A. P = 0.010

B. P = 0.005

C. P = 0.001

D. P = 0.000025

 

7. Reliability prediction is :

A. A one time estimation process.

B. A continuous process starting with paper predictions.

C. More important than reliability attained in the field.

D. A popular method as simulation theory.

 

8. Ideally for a FRACAS to be effective, how many failures should be allowed to pass before corrective action is to be undertaken ?

A. First occurrence of a failure mode.

B. Second occurrence of a failure mode.

C. Third occurrence of a failure mode.

D. Fourth occurrence of a failure mode.

 

9. All of the following Boolean algebra expression are incorrect EXCEPT ?

A. 1 + 1 = 2

B. 1 – 1 = 1

C. 1 – 0 = 0

D. 1 + 0 = 1

 

10. What is the MOST accurate method to verify that the maintainability requirement of a design is being met ?

A. By analysis of the design.

B. By performing maintainability prediction.

C. By thorough design reviews.

D. By demonstration at the customer’s facility.

 

Picture © B. Poncelet https://bennyponcelet.wordpress.com

Posted in General

RECORDED WEBINAR: NEXT GENERATION SOFTWARE RELIABILITY PREDICTION USING CAUSAL LEARNING by Robert Stoddard

Software reliability practice continues to evolve from a early focus on the modeling of software test failures for reliability estimation to the modeling of pre-test activities and software attributes for reliability prediction.
The speaker believes the next major evolutionary step in software reliability research and practice will come with the application of causal learning.
Causal learning has become a practical and exciting field rooted in matching methods employed long before Ronald Fisher created Designed Experimental methods in the 1930s and 1940s.
This webinar will share the recently matured landscape of causal learning consisting of causal discovery and causal estimation.
A brief description of causal methods, algorithms and modern publications will be shared along with recommendations on how reliability engineers might pursue learning and adopting causal learning.

Posted in General

Recorded webinar: BASIC QUANTITATIVE RMA FOR THE PRACTITIONER by Tim Adams

RMA stands for Reliability, Maintainability and Availability.

This presentation targets the practitioner working basic quantitative Reliability, Maintainability, and Availability (RMA).
The presentation’s sequence is:
1. RMA concepts are described, and five central questions in RMA are stated to describe basic competencies in probabilistic RMA.
2. Each central question is illustrated with an example, and each example is worked in Microsoft Excel.
3. Three of the questions and associated examples pertain to forecasting reliability for the following scenarios:
a. New item with no downtime
b. Used item with no downtime
c. New item with scheduled downtime for idealized preventive maintenance.
4. The three mentioned reliability cases are compared as a means to summarize principles pertaining to when break-in and preventive maintenance provide a benefit to the reliability measure.
5. A process that transforms data to a math model for reliability and maintainability is described.
6. Sources for life data and tips for making a data collection program are summarized.

Posted in General

Reliability analysis using Reliability Block Diagram (RBD) – WEBINAR SLIDES

On April 12, 2018 Frank Thede presented “Reliability analysis using Reliability Block Diagram (RBD)”
Below a link with the slides of this webinar.

Frank Thede brings 20+ years experience in all aspects of asset management, from capital project management, maintenance management and reliability improvement, to the end of life replacement programs. Frank is the Principal Reliability Engineering at Reliability Works. He has worked on a large variety of projects in Power Generation and Transmission, Oil and Gas, Aluminum, Marine, Transportation and Telecommunications. Frank’s extensive background in electrical engineering, combined with his specialization in reliability and maintenance management provides him with the necessary skill set and experience to effectively manage any group of physical assets.

Reliability analysis using Reliability Block Diagram (RBD).

The recorded webinar will be released later.

Posted in General

Demystifying the common misconceptions about Reliability Centered Maintenance (RCM) – WEBINAR SLIDES

On March 8, 2018 Nancy Regan presented “Demystifying the common misconceptions about Reliability Centered Maintenance (RCM)”

Below a link with the slides.

Nancy Regan is the founder of RCMTrainingOnline.com. Nancy is a graduate of Embry-Riddle Aeronautical University with a B.S. degree in Aerospace Engineering. As a U.S. Navy civilian employee for seven years, she completed Naval Aviation Maintenance Officer School. She then became Team Leader for RCM at the Naval Air Warfare Center, Aircraft Division, Lakehurst, NJ, where she instituted the RCM Program on Naval Aviation Common Support Equipment. In 2001 she left government service and founded The Force, Inc. Nancy has over 20 years’ experience of hands-on practice facilitating RCM analyses, conducting RCM training, and assisting her clients in implementing RCM programs on aircraft, manufacturing equipment, and all kinds of equipment in between. Amongst the many projects she has facilitated is the CH-47 Chinook Helicopter, the US Army’s heavy-lift helicopter. Nancy holds U.S. and foreign patents on a process for marking parts that she developed using her RCM experience. She is the author of The RCM Solution, A Practical Guide to Starting and Maintaining a Successful RCM Program. Nancy is dedicated to bringing affordable and accessible RCM training to the Maintenance and Reliability community. She resides in Huntsville, Alabama with her husband, Dennis.

Demystifying the common misconceptions about Reliability Centered Maintenance (RCM)

The recorded webinar:

Posted in General

The Reliability Division has received the Bronze award recognition for 2017

We are pleased to announce that the Reliability Division has received the Bronze award recognition for 2017. This level was achieved by meeting the good standing requirements as well as meeting or exceeding the Increase in Growth and Retention %metrics.

Posted in General

An overview of the Reliability Practice – Webinar slides

Earlier this week Norma Antunano presented the webinar: An overview of the Reliability Practice

Below a link with the slides.

Norma has worked in the technology industry in different Quality and Reliability roles and is passionate about Continuous Improvement; she has led and mentored diversity of technical, business improvement and innovation projects at Medtronic, Honeywell, Broadcom and Hewlett Packard from scoping through completion with successful results. Norma has been Malcolm Baldrige National Quality Award Examiner for over four years, facilitates diversity of courses including Probability and Statistics for Business graduate students, and has authored papers for IPC, ECTC and ASQ. She is Systems Engineer, holds a Master in Engineering Sciences, MBA in International Management and PhD in Engineering Philosophy. Norma is region 14A ASQ Reliability Councilor and Section 1414 Education Chair, is IEEE Senior member, and is also ASQ certified CQE, CSQE, CRE and SSBB.

An overview of the Reliability Practice

The recorded webinar will be released later.

 

Posted in General

Update ASQ Reliability & Risk Division

Hello ASQ Reliability & Risk Division Members and Volunteers,

As most of you are probably aware of by now, ASQ is undergoing a transformation that may significantly alter how technical communities such as the ASQ Reliability & Risk Division operate and our future state.

The ASQ Reliability & Risk Division Management Committee met recently to discuss our future in the context of this transformation and you may see some changes as a result. Our end goal is to preserve ASQ member value and the value that we bring to the reliability and risk professional community at large. And we still intend for our division to be subject matter experts and thought leaders through our webinar offerings, conference sponsorship and participation, Body of Knowledge development, and outreach.

Due to the changing IT landscape, you will see changes in the www.asqrd.org website and access to certain content to make it simple and easier to manage in the near term and as we prepare for other aspects of ASQ transformation. Some content may be deactivated in the near term as we sort out access rights since the content is under copyright by other organizations. We will not lose any of this content since we will maintain redundant storage of it but it just may be temporarily unavailable.

Specifically, past proceedings and webinar presentations will no longer be available on our website. Our archived webinar videos will still be available without a required login at Vimeo. All our videos can be accessed at https://vimeo.com/album/5072285 or by using the menu link above. Additionally, if you wish to be added to the future webinar mailing list please send us an email at webmaster @ asqrrd .org (remove the spaces).

We will continue to be a strong presence at the RAMS Symposium and will be considering collaboration with other ASQ Division sponsored conferences and external conferences to bring our message, content, and training to a wider audience with more efficiencies and impact.

Of course, to continue to provide the member value that we do, proper funding of our efforts needs to be made available since even a non-profit, member led, volunteer society and its divisions rely on funding and wise management of those funds to continue to be viable. We do have concerns about the funding impact to our division and other divisions in the Technical Communities Council that ASQ transformation may bring and we will be working to ensure that the proper amounts and mechanisms to protect those funds be put in place so that our vital mission can continue.

Best Regards,
Dan Burrows
Chair – ASQ Reliability & Risk Division

Posted in General

TECH SPOT: Sample CRE questions

800-11-5321

1. Which answer BEST describes the events and operating conditions an item experiences from mission initiation to completion? The events may include the research and development phase, product manufacturing, warehousing, and so on, to mission completion.

A. Operational readiness; B. Mission profile; C. Design adequacy; D. Mission reliability

2. What is System Effectiveness, if Operational Readiness is 0.89, Design Adequacy is 95%, Availability is 99%, Maintainability is 0.93, and Mission Reliability is 0.99? Ebeling, p 149

A. 0.763; B. 0.881; C. 0.837; D. 0.820

3. Which of the following functions are normally accepted reliability engineering tools?

I. Failure probability density function; II. Failure rate function; III. Reliability function; IV. Conditional reliability function; V. Mean life function.

A. I and II only; B. I, II and III only; C. I, II, III and IV only; D. I, II, III, IV and V

4. An airline maintains a fleet of 4-engine aircraft. Its maintenance records show that on the average an engine fails 3 times in 10,000 operating hours with normal preventive maintenance. What is the Poisson distributed probability that 2 or more engines on an aircraft will fail during a typical flying period of 8 hours?

A. 0.000034; B. 0.0000034; C. 0.0000029; D. 0.000029

5. For the exponential model, the reliability at mean time to failure is about:

A. 37 percent; B. 50 percent; C. 67 percent; D. 73 percent

6. A plastics plant operates 8 extruders producing plastic film. Production volume requirements cannot be met if less than 6 extruders are operating. There is a .30 probability that a machine stopping malfunction will occur. What is the probability that 6 extruders can remain operating throughout the day?

A. 0.5783; B. 0.4482; C. 0.5518; D. 0.8059

7. A system is made up of four independent components in series each having a failure rate of .005 failures per hour. If time to failure is exponential, then the reliability of the system at 10 hours is:

A. 0.8187; B. 0.8860; C. 0.9512; D. 0.9802

8. What is the reliability of this system?

Screenshot 2018-03-17 at 16.59.34

Where component reliabilities are: A. 0.80; B. 0.95; C. 0.82; D. 0.85; E. 0.75

A. 0.10; B. 0.90; C. 0.95; D. 0.04

9. Which method is used to predict new device reliability during its early design stage?

A. Burn-in method; B. Part stress analysis method; C. Parts count method; D. Accelerated testing method

10. Which of the following forms of reliability data will BEST provide valuable information on product usage and reliability?

A. In-house test results. B. Independent lab results. C. Field support data. D. Quality control data.

Picture © B. Poncelet https://bennyponcelet.wordpress.com

Posted in General