| Course | OPS 385 Lean Six Sigma and Process Management (OPS/385) |
|---|---|
| Week | 3 |
| Paper type | DMAIC measure and analyze phases |
| Length | about 1,069 words, 4 double-spaced pages plus title page and references |
| Format | APA 7 student paper |
| School | University of Phoenix |
| Program | BS in Business |
| Updated | October 2026 |
Free sample paper for OPS 385 Week 3
Is It the Bench, the Rebuilder or the Regulator? Measuring and Analyzing Alternator Test Failure Data
[Student Name]
University of Phoenix
OPS/385: Lean Six Sigma and Process Management
Week 3 Assignment
[Instructor Name]
[Date]
Delta Rebuild Industries, its data, test results and statistics are composites written for a model paper.
Delta Rebuild Industries, the composite Memphis remanufacturer in this course, chartered a project in Week 2 to cut alternator bench test failures from about 7 percent to 2 percent or less. The define phase set requirements and scope. This paper reports the measure and analyze phases: confirming that the test bench can be trusted, establishing the baseline and finding causes the data support.
Can the Bench Be Trusted?
Before blaming the process, the team checked the instrument. A gauge repeatability and reproducibility study used ten alternators spanning the range of outputs, measured twice each by three test technicians on the bench, with the order randomized. Montgomery and Runger (1993) described how such studies separate measurement variation into repeatability, from the instrument, and reproducibility, from the people using it, so that it can be compared with the variation in the parts or with the specification. The study found that measurement variation used about 8 percent of the voltage tolerance and about 9 percent of the output current tolerance, both below the commonly used 10 percent level for an acceptable system. The bench is fit to use.
The Data Collection Plan
The team wrote a one-page plan before collecting anything. For each unit, the test technician would record the result, the failure mode from a fixed list of six and the unit's serial number. The rebuild traveler, a paper form that follows each unit, already recorded the rebuilder and shift; the team added two check boxes, one for the regulator supplier and one for whether the slip rings were machined, and a line for the brush lot number. The quality engineer reviewed travelers daily for missing entries during the first week, when about one in eight was incomplete, and coached rebuilders until the rate fell below one in fifty. Planning these details in advance meant that when a hypothesis came up later, the data to test it were already there.
Baseline
Over six weeks, the team recorded every rebuilt heavy-duty alternator's test result with its failure mode, rebuilder, shift, regulator supplier and whether the slip rings were machined. Of 1,260 units, 93 failed, a rate of about 7.4 percent, or about 74,000 failures per million units. That corresponds to roughly a 2.9 sigma process on the conventional scale, which allows for a 1.5 sigma long-term shift.
Failure Modes
A Pareto chart of the 93 failures by mode showed low output at idle, 41 percent; voltage above 14.6 volts, 23 percent; bearing or diode noise, 19 percent; and other modes, 17 percent. Low output and high voltage together account for almost two-thirds of failures.
Generating Hypotheses
A cause-and-effect session with rebuilders, test technicians and the quality engineer listed possible causes under six headings, following the approach Ishikawa (1986) popularized. The team selected five hypotheses to test with data: rebuilder skill, shift, regulator supplier, slip ring condition and brush supplier. Each was chosen because someone with direct knowledge had seen it matter.
A fishbone diagram is a list of suspects, not a verdict.
Stratification
Rebuilders: failure rates for the seven rebuilders ranged from 6.1 to 8.4 percent, a spread consistent with chance given their volumes. Shift: day shift 7.2 percent, second shift 7.7 percent, also no meaningful difference. Brush supplier: no difference. Regulator supplier: units built with regulators from supplier A failed at 3.9 percent, 27 of 700; units with regulators from supplier B, introduced eight months ago to save about $6 a unit, failed at 11.8 percent, 66 of 560.
Testing the Regulator Difference
A chi-square test compared the failure proportions for the two suppliers. If supplier made no difference, about 51.7 failures would be expected among supplier A units and 41.3 among supplier B units. The observed counts were 27 and 66. The chi-square statistic, built by squaring each gap between an observed and an expected count, scaling it by the expected count and adding the four results, comes to about 28.7. With a single degree of freedom, a statistic that large has a p-value far below 0.001. That means a difference this large would be extremely unlikely if supplier made no difference. It does not by itself prove the regulators are faulty, but combined with the failure modes, high-voltage failures were almost all supplier B units, it strongly supports regulator quality as a cause.
The Slip Ring Finding
Low output at idle was spread across both suppliers. The team examined machining records: slip rings are machined only when they look worn, and rebuilders judged by eye. Units whose slip rings were not machined failed for low output at about 4.6 percent; those machined failed at about 1.2 percent. Measuring slip ring runout on 60 unmachined units found that 22 exceeded the manufacturer's specification despite looking acceptable. Visual judgment was letting out-of-specification rings through, causing poor brush contact and low output.
Checking the Supplier Finding From Another Angle
Because a supplier change carries cost and contract implications, the team sought a second line of evidence. It bench-tested 30 new regulators from each supplier, straight from the box, on a regulator tester. Eight of the supplier B units regulated above 14.6 volts at high load, against one of the supplier A units. The independent test, done before the units entered any alternator, points to the part itself rather than to how rebuilders installed it.
What Was Ruled Out
Rebuilder skill, shift and brush supplier were tested and did not explain the failures. Ruling them out matters: the team had expected rebuilder training to be the answer, and that effort would have been wasted. Antony (2004) noted that one strength of Six Sigma is its insistence on data-driven decisions, which helps organizations avoid acting on assumptions.
Root Causes for the Improve Phase
Two root causes are supported by data: regulators from supplier B, linked mainly to high-voltage failures, and slip rings passed by visual inspection without measurement, linked mainly to low output. Together they account for an estimated 70 percent of failures.
Conclusion
The measure and analyze phases confirmed the test bench is reliable, set a baseline of about 7.4 percent failures and tested five suspected causes. Stratification and a chi-square test pointed strongly to regulator supplier B, and measurement of slip rings revealed that visual inspection misses out-of-specification rings. Three suspected causes were ruled out. The improve phase can now target the causes the data support.
References
Antony, J. (2004). Some pros and cons of six sigma: An academic perspective. The TQM Magazine, 16(4), 303-306. https://doi.org/10.1108/09544780410541945
Ishikawa, K. (1986). Guide to quality control (2nd ed.). Asian Productivity Organization.
Montgomery, D. C., & Runger, G. C. (1993). Gauge capability and designed experiments. Part I: Basic methods. Quality Engineering, 6(1), 115-135. https://doi.org/10.1080/08982119308918710
What the OPS 385 Week 3 instructions ask
In the third OPS 385 assignment, students typically complete the measure and analyze phases of a Lean Six Sigma project. Prompts may ask for a data collection plan, measurement system analysis such as a gauge repeatability and reproducibility study, baseline performance and sigma level or yield, graphical analysis such as Pareto charts, histograms and stratification, cause-and-effect analysis and at least one statistical test of a suspected cause. Use realistic data, show calculations and explain what each result means for the project. Support the analysis with quality and statistics references cited in APA, and avoid declaring causes the data do not support.
How this OPS 385 Week 3 example is built
Here the measure phase begins with the instrument rather than the process: a gauge study on the alternator test bench shows that measurement variation is small enough to trust. Six weeks of data on 1,260 rebuilt alternators give a baseline failure rate of about 7.4 percent. A Pareto chart of failure modes puts low output at idle first, followed by high voltage and bearing noise. Stratification finds no meaningful difference between rebuilders or shifts but a large difference between two regulator suppliers. A chi-square test confirms that the supplier difference is very unlikely to be chance. A second analysis links low output to slip rings that skipped machining because they looked acceptable. The paper ends with two verified root causes.
OPS 385 Week 3 grading rubric: where the points go
Graders expect the measure and analyze phases to be done in order and with evidence. Strong papers validate the measurement system before using its data, present a baseline with a clear calculation and stratify the data in ways linked to suspected causes. Credit goes to correct use and interpretation of at least one statistical test, to cause-and-effect analysis that generates hypotheses and to conclusions limited to what the data show. Papers that report negative findings, such as a suspected cause that did not hold up, show sound reasoning. Clear charts or tables, accurate statistics and correctly formatted APA references complete the strongest submissions.
OPS 385 Week 3 help: mistakes to avoid
Many papers skip measurement system analysis and trust whatever the gauge says. If the instrument is noisy, every later conclusion is suspect. Check it first. Another frequent error is drawing a fishbone diagram and treating every branch as a cause; the diagram lists possibilities that data must test. Students also misinterpret statistical results, for example treating a p-value as the probability that the hypothesis is true. Explain it correctly. Some papers stratify by every variable available without a reason; stratify by the factors the team suspects. Finally, report what you ruled out, since that saves the improve phase from wasted effort and shows the grader your reasoning. A tutor can walk through your test calculations with you.
Related OPS 385 sample papers
Other OPS 385 week samples
- OPS 385 Week 1: Value, Waste and Lean Thinking
- OPS 385 Week 2: Defining a DMAIC Project
- OPS 385 Week 4: Capability and Control Charts
- OPS 385 Week 5: Improving and Sustaining the Process
More BS in Business sample papers
- MKT 498 Week 3: Media Selection and Planning
- OPS 330 Week 3: Inventory and Supply Chain Control
- PM 300 Week 3: Governance and Project Environments
- PM 310 Week 3: Development Approach and Life Cycle
OPS 385 Week 3 questions, answered
What does OPS 385 Week 3 usually cover?
It usually covers the measure and analyze phases of DMAIC: data collection, measurement system analysis, baseline performance, Pareto and stratification, cause-and-effect analysis and statistical tests of suspected causes.
Where can I find a free OPS 385 Week 3 sample paper?
The Week 3 paper above measures and analyzes alternator test failure data with a gauge study and chi-square test, and it is free to read.
What is a gauge R&R study?
A study that measures how much of the variation in measurements comes from the measuring instrument and the people using it, compared with the variation in the parts being measured.
What does a chi-square test show in Six Sigma?
Whether the proportions of an outcome, such as failures, differ between groups, such as suppliers, by more than chance alone would likely produce.
What is stratification in data analysis?
Breaking data into groups, such as by shift, supplier or operator, to see whether a problem is concentrated in one group, which points toward its cause.
Write yours, or have the desk draft it
This paper is an original model document written by our desk, not a submitted student paper and not an official University of Phoenix document. Read it for the moves, then write your own to the instructions in your classroom. If you want one built to your exact prompt and rubric, the first custom sample is free and arrives in 24 to 48 hours.
Request this one custom, free · All OPS 385 week samples · All courses