| Course | MT438 Analytics in the Digital Supply Chain |
|---|---|
| Unit | Unit 8 |
| Paper type | Root cause analysis |
| Length | About 1,100 words, 7 pages double-spaced |
| Format | APA 7 student paper |
| School | Purdue University Global |
| Program | BS in Business Administration |
| Updated | October 2026 |
Free sample paper for MT438 Unit 8
Not the Downtime: An MT438 Unit 8 Root Cause Analysis of a Farm-Store Chain's Falling Fill Rate
[Student Name]
Purdue University Global
MT438: Analytics in the Digital Supply Chain
Unit 8 Root Cause Analysis
[Instructor Name]
[Date]
Oakhurst Lawn and Garden, the farm-store chain, the plants and every figure are composite; the sources cited are real.
The Problem, Stated Precisely
Between March 9 and May 1, 2026, Oakhurst Lawn and Garden cut 526 of the 4,619 order lines placed by its farm-store customer, a regional chain of 140 rural stores. That is a cut rate of 11.4 percent and a line fill rate of 88.6 percent. In the previous eight weeks, January 12 to March 6, the chain's cut rate was 2.8 percent and its fill rate 97.2. The farm-store chain does not charge fines, but its buyer has said that a second season like this one would move part of its seed business to a competitor. The decline is stated with dates and counts because a vague complaint that "fill rate is down" cannot be tested.
The Assumed Cause
Within a week of the first complaints, the explanation in every meeting was downtime. Decatur's bagging line lost thirty-one hours in March to a sealing failure, and the plant manager's report linked the downtime to the cuts. The analysis treats that explanation as a hypothesis rather than a fact. Okes (2019) recommends writing down what should be true if a suspected cause is the real one, then checking each consequence before accepting it; the three predictions below follow that method.
Prediction 1: Production Should Have Fallen Short
If downtime caused the cuts, Decatur and Macon together should have produced less than planned. They did not. Production over the eight weeks reached 98.6 percent of plan for the products the chain buys. Decatur recovered its lost hours with two Saturday shifts in late March. Missing production of 1.4 percent cannot explain an increase in the cut rate from 2.8 to 11.4 percent.
Prediction 2: Cuts Should Cluster at Decatur
If the bagging line were the cause, the decline should center on the plant where it failed. It did not. Line fill on orders shipped from Decatur fell from 96.9 to 88.2 percent. Line fill on orders shipped from Macon, where nothing broke, fell from 97.6 to 89.1. Two plants with different equipment, different crews and different shipping lanes declining by nearly identical amounts point to a cause the plants share, not one local to either.
Prediction 3: Cut Products Should Have Been Out of Stock
If production fell short, cut lines should involve products that were not in the warehouse when the order arrived. For 373 of the 526 cut lines, 71 percent, the product was physically in the building at the moment the order was entered. Stock was present; it simply was not available to this customer.
Seven in ten cut lines involved product sitting in the warehouse when the order arrived.
Widening the Search
With downtime ruled out, the analysis listed alternative causes in a fishbone diagram whose branches cover ordering behavior, inventory allocation, production and transport. Andersen and Fagerhaug (2006) describe the diagram as a way to make sure no category of cause is skipped, not as evidence in itself, and it is used that way here. Each branch suggested a way to slice the cut lines.
Slices That Showed Nothing
Cuts were sorted by product family, by carrier and by day of the week. None showed a pattern. Seed, fertilizer and weed control lines were cut in roughly the proportion they were ordered. The three carriers serving the chain had cut rates within a point of each other. Cuts spread evenly across days of the week. These slices are reported because they show the finding below was not the first breakdown tried; a reader who sees only the revealing slice cannot tell whether the search was wide or stopped early. Wang et al. (2016) observe that most logistics analytics stays descriptive, reporting what happened; slicing the same records against stated hypotheses is what moves the work toward explaining why.
The Slice That Did
Sorting cut lines by order lead time, the days between order entry and scheduled shipment, revealed the pattern. Lines ordered seven or more days ahead, 1,718 of them, were filled at 97.0 percent. Lines ordered four to six days ahead, 1,007, were filled at 93.5 percent. Lines ordered less than four days ahead, 1,894, were filled at only 78.4 percent, and they account for 409 of the 526 cuts. Before March, short-notice lines were 6 percent of the chain's orders. In the problem period they were 41 percent.
Why Lead Times Shortened
On March 2, the chain moved from weekly to twice-weekly ordering, a change its buyer described as a way to hold less stock in its rural stores. Each order now covers fewer days, so it is placed closer to when the stores need product. Nothing in Oakhurst's planning changed to match. The chain's order quantities fell by about half per order, and the lead time between order and shipment fell with them.
Why Short-Notice Orders Lose
The second half of the cause lies in Oakhurst's planning system. Each night, the system reserves inventory against the home-improvement account's forecast for the next five days, so that the largest account's appointments are protected. A farm-chain order placed more than five days ahead is entered before the reservation covers its ship date and gets first claim on stock. A short-notice order arrives after the reservation and finds product present but spoken for. This rule was harmless while the farm chain ordered weekly and far ahead. Twice-weekly ordering pushed four in ten of its lines into the reserved window.
The root cause is therefore a combination: a customer's change in ordering, which Oakhurst did not notice, meeting an allocation rule that ranks customers by when they order rather than by agreed priority.
Verification Plan
A cause is not established until changing it changes the result. The analysis proposes a four-week trial starting June 1. For the products the farm chain buys, the reservation window will be shortened from five days to two. If the diagnosis is correct, fill on the chain's short-notice lines should rise from 78.4 percent toward the roughly 95 percent achieved on longer-notice lines, while fill to the home-improvement account should fall by no more than half a point. If short-notice fill does not move, the analysis is wrong and the search resumes. Either way, the farm chain's buyer should be told about the trial and asked whether twice-weekly ordering is permanent.
References
Andersen, B., & Fagerhaug, T. (2006). Root cause analysis: Simplified tools and techniques (2nd ed.). ASQ Quality Press.
Okes, D. (2019). Root cause analysis: The core of problem solving and corrective action (2nd ed.). ASQ Quality Press.
Wang, G., Gunasekaran, A., Ngai, E. W. T., & Papadopoulos, T. (2016). Big data analytics in logistics and supply chain management: Certain investigations for research and applications. International Journal of Production Economics, 176, 98-110. https://doi.org/10.1016/j.ijpe.2016.03.014
What the MT438 Unit 8 assignment asks for
Unit 8 of MT438 expects a root cause analysis of a supply chain performance problem. State the problem precisely, with counts, dates and the comparison period, because a vague decline cannot be tested. If an explanation is already accepted, treat it as a hypothesis and write down what should be true if it were correct. Test each prediction with data. A cause-and-effect diagram helps list alternatives across categories, but it organizes the search rather than proving anything. Report the breakdowns that revealed nothing along with the one that did, so the reader can see how wide the search was. When a cause emerges, explain the mechanism that links it to the outcome. Close with a way to verify the cause, usually a trial with an expected result. Six pages with a few tables is a common scope.
How this MT438 Unit 8 example is built
The Oakhurst sample opens with 526 of 4,619 lines cut between March 9 and May 1 against a 2.8 percent cut rate earlier. The downtime explanation is stated as a hypothesis, citing Okes. Three short sections test its predictions on production, plant and stock, and a highlighted line notes that seven in ten cut lines had product in the warehouse. A fishbone diagram widens the search, with Andersen and Fagerhaug on its proper use. A section reports the slices that showed nothing, with Wang and colleagues on moving past descriptive analytics. The lead-time slice follows with fill at 97.0, 93.5 and 78.4 percent. The mechanism comes in two parts: the chain's move to twice-weekly ordering on March 2 and a five-day reservation for another account. A four-week trial with a two-day window closes the paper as verification.
Reading the MT438 Unit 8 grading rubric
Rubrics in this unit reward testing over storytelling. A root cause analysis that accepts the first explanation and documents it collects few analysis marks; one that states predictions and checks them earns most of them. Precision in the problem statement is credited because it makes testing possible. Graders look for a cause-and-effect diagram used to widen the search, not offered as evidence. Reporting dead-end slices shows the finding was not the first breakdown tried, and many instructors mark this explicitly. A mechanism linking cause to outcome is expected; a correlation by itself is not a root cause. Verification is a frequent marking point, and a trial with a stated expected result shows the student understands the difference between a likely cause and a proven one. Tables should be labeled and referenced. Purdue Global expects quality and analytics sources cited in APA style.
MT438 Unit 8 help: mistakes that cost points
The commonest Unit 8 mistake is confirming the explanation everyone already believed, with data selected to fit it. Another is presenting a fishbone diagram full of possible causes and then choosing one without testing any. Students often report only the slice that showed a pattern, which leaves the grader unsure whether it was found or fished for. Problem statements are frequently vague, saying performance declined without counts or dates. Some analyses stop at the first factor that correlates with the problem and never explain why it would cause it. Verification is often missing entirely or reduced to monitoring going forward. Before submitting, write the predictions your chosen cause makes and check each one, just as you did for the cause you rejected. If any prediction fails, you have more work to do before calling it the root cause.
Get an MT438 Unit 8 example written to your instructions
Send what the Unit 8 case says about the declining measure, the explanation people first gave, and any data behind both, plus your assignment instructions and rubric. What we return is a composite analysis that tests the first explanation's predictions, shows every slice tried and proposes a verification. The opening custom sample carries no fee and lands within 24-48h. The paper above is an original model document written by our desk, not a submitted student paper and not an official Purdue University Global document.
More MT438 papers and BS in Business Administration samples
- MT438 Unit 1: Supply Chain Data Inventory Post
- MT438 Unit 2: Order Data Quality Audit
- MT438 Unit 3: Supply Chain Metrics Exercise
- MT438 Unit 4: Operations Dashboard Critique
- MT438 Unit 5: Grass Seed Forecast Accuracy
- MT438 Unit 6: Production Cut Seminar Reflection
- MT438 Unit 7: Shipment Tracking Technology Brief
- MT438 Unit 9: Predictive Model Evaluation
- MT438 Unit 10: Analytics Recommendation Memo
- MT437 Unit 8: Cycle Count Plan
- MT436 Unit 8: Contract Clause Review
- MT435 Unit 8: Control Chart Analysis
- MT434 Unit 8: Delivery Service Tier Proposal
MT438 Unit 8 questions, answered
Do I have to use a fishbone diagram?
Many sections expect one, and it helps organize candidate causes across categories. It is a brainstorming tool, though, not evidence. The analysis earns its credit in the tests that follow, where each candidate's predictions are checked against data. If your prompt names another method, such as five whys, use it the same way: as structure for testing, not as the conclusion.
How do I test whether an assumed cause is right?
Write down what should be true if it were the cause: which locations, products or periods should be affected, and what other measures should have moved. Then check each prediction. A cause that fails its own predictions can be set aside even if it seemed obvious, and saying so plainly is one of the strongest moves available.
Why include analyses that found nothing?
Because they show the finding was not the only slice tried. If you report only the breakdown that revealed a pattern, a reader cannot tell whether you searched widely or stopped at the first interesting result. A short table of dead ends, with one line each, costs little space and makes the real finding far more credible.
Where can I find a free MT438 Unit 8 sample paper?
This page prints the complete MT438 Unit 8 analysis, which rejects a downtime explanation and traces a farm-store chain's cut lines to short order lead times.
What does the MT438 Unit 8 sample propose to prove its cause?
A four-week trial shortening the stock reservation window from five days to two, expecting short-notice fill to rise from 78.4 percent toward about 95.