GM504 Unit 5 Delivery Correction Measurement Plan Example

Course	GM504 Organizational Excellence and Change
Unit	Unit 5
Paper type	Measurement and baseline plan
Length	About 880 words, 6 pages double-spaced
Format	APA 7 student paper
School	Purdue University Global
Program	MS in Management
Updated	October 2026

Free sample paper for GM504 Unit 5

First Agree What an Error Is: A Unit 5 Measurement and Baseline Plan for Delivery Corrections

[Student Name]

Purdue University Global

GM504: Organizational Excellence and Change

Unit 5 Measurement and Baseline Plan

[Instructor Name]

[Date]

Ridgeway Building Supply, its clerks and every count are composite and illustrative; the methods and studies cited are real.

What this part is doingOpening on the definition problem tells the reader the plan treats measurement itself as something that must be proven before results can be trusted.

Why the Plan Comes First

The temporary loading checker from two winters ago seemed to work because corrections fell, but nobody had written down what counted as a correction, and different branches counted differently. One branch counted every credit memo, including price adjustments; another counted only returns. Without a shared definition, the project would have no way to tell a real improvement from a change in how errors were recorded.

The Primary Measure

Name: correction rate.

Operational definition: the number of delivery orders that generate a redelivery or a credit memo for a delivery error within ten days of shipment, divided by all delivery orders shipped that week. A delivery error is a wrong item, short quantity, damaged item, missing add-on or other fault in what arrived. Credit memos for pricing disputes, customer changes of mind and returns of correct items are excluded.

Data source: the billing system's credit memo log and the dispatch system's redelivery records, joined by order number.

Collected by: the billing clerk at each branch, who codes each correction by reason.

Frequency: weekly, with the week defined by shipment date.

Display: a run chart by week for the whole company, with a separate line for each branch.

Supporting Measures

Bin-location accuracy. Each week the inventory control lead checks fifty randomly selected bins and records whether the product in the bin matches the system's record. Accuracy is the share that match. This measure tracks the root cause identified in Unit 4, so a change in it should precede any change in wrong-item corrections.

Defects per million order lines. Because orders vary in size, a large order has more chances for an error than a small one. Dividing corrected lines by all lines shipped, scaled to a million, gives a figure that can be compared across branches of different size and over time. Six Sigma practice uses this kind of measure because it ties the goal to the opportunities for error rather than to a count that grows with volume (Linderman et al., 2003).

Testing the Measurement System

A measure is only useful if the people collecting it apply the definition the same way. An attribute agreement test checks this by giving several appraisers the same items and comparing their judgments (Automotive Industry Action Group, 2010).

Three billing clerks, one from each of three branches, independently coded the same forty credit memos by reason. All three agreed on 29 of the forty, 72.5 percent. Most disagreements involved two pairs of categories: damaged versus wrong item, when a product arrived broken and was replaced with a different one, and short quantity versus missing add-on, when a contractor phoned in an extra item that did not arrive.

The clerks agreed on only 29 of forty memos at first, which meant any change in the rate could have come from how they coded rather than what went wrong.

What this part is doingReporting the low first agreement, not hiding it, shows the plan tested its own measure honestly.

The definitions were revised with examples for each disputed case. A damaged item is coded as damage even if a different item is sent as a replacement; an add-on is coded as missing only if it was recorded on the order before shipment. The clerks then coded a fresh set of forty memos and agreed on 37, 92.5 percent, which the project accepted as adequate for a weekly operational measure.

The Baseline

The baseline covers twelve weeks before any change. Across those weeks, the company shipped about 7,200 delivery orders and recorded 418 corrections, a correction rate of 5.8 percent. Weekly rates ranged from 5.1 to 6.6 percent.

Plotted on a proportion control chart, which sets limits based on each week's order count, no week falls outside the limits and no run of points suggests a shift (Montgomery, 2020). The process is stable at a rate that is too high: its variation comes from common causes within the system, not from unusual events. That matters for the project, because a stable process will not improve by addressing particular bad weeks; only a redesign of how the work is done will move the rate.

What this part is doingEstablishing that the baseline is stable tells the project it must change the system rather than chase individual bad weeks.

Baseline by Category

The twelve weeks confirm the pattern from the single month analyzed in Unit 4. Wrong item accounts for about 37 percent of corrections, short quantity for about 30 percent, and damage, missing add-ons and other reasons for the rest. Bin-location accuracy averaged 84.5 percent across the twelve weekly samples, consistent with the 200-bin audit.

The Goal

The project charter sets a goal of reducing the correction rate to 3.5 percent within six months of implementation. Reaching it would require cutting corrections by about 40 percent. That target is ambitious but grounded in the root cause analysis: if most wrong-item corrections come from mismatched bins, fixing location accuracy could remove a large share of the largest category.

Ownership

Until the project ends, the project lead owns the measurement plan. The control plan in Unit 9 will transfer ownership of each measure to the person responsible for the process it tracks.

References

Automotive Industry Action Group. (2010). Measurement systems analysis (4th ed.). AIAG.

Linderman, K., Schroeder, R. G., Zaheer, S., & Choo, A. S. (2003). Six Sigma: A goal-theoretic perspective. Journal of Operations Management, 21(2), 193-203. https://doi.org/10.1016/S0272-6963(02)00087-6

Montgomery, D. C. (2020). Introduction to statistical quality control (8th ed.). Wiley.

GM504 Unit 5 instructions, in plain terms

GM504's fifth unit asks for a measurement plan and a baseline before any change is made. For each measure, state its name, an operational definition precise enough that two people would reach the same answer, the data source, who collects it, how often and how it will be displayed. Include a primary measure and supporting measures, especially one that tracks the suspected root cause. Then test the measurement system itself. Where people code or classify records, an attribute agreement check is the simplest test, and low agreement should lead to revised definitions and a retest. Gather enough weeks of baseline data to reveal how much the rate normally moves, plot it on an appropriate control chart and say whether the process is stable. Restate the goal and its basis. Assign ownership of the measures for the life of the project. Expect about five pages built around a measurement plan table, citing measurement and statistical sources in APA 7.

How this GM504 Unit 5 example is built

The plan opens by explaining that branches counted corrections differently, so a fall in the rate could have been a change in counting. The primary measure follows with its definition, exclusions, sources, collector, frequency and run-chart display. Bin-location accuracy from a weekly fifty-bin sample tracks the root cause, and defects per million lines allows comparison across branches, with Linderman and colleagues cited on Six Sigma goals. An attribute agreement test, drawing on the AIAG manual, reports 29 of forty, and the highlighted sentence explains why that mattered. Revised rules for damage and add-ons raise agreement to 37 of forty. The baseline section reports 418 corrections over twelve weeks, weekly rates of 5.1 to 6.6 percent and a stable proportion chart, citing Montgomery. The category split, a 3.5 percent goal and project-lead ownership close the plan.

Where the points sit in the GM504 Unit 5 rubric

Most of the credit on this GM504 plan comes from definitions precise enough to use and a measurement system shown to be reliable. Instructors check each measure for a clear operational definition, including what is excluded. Supporting measures should be tied to the analysis, especially the root cause. An attribute agreement test earns significant credit, and reporting a low first result honestly, followed by revision and retest, shows mature practice. The baseline should be long enough to show variation and displayed on a chart suited to the data; stating whether the process is stable is expected. A goal grounded in the root cause analysis, and clear ownership, complete the plan. The AIAG measurement systems manual, Montgomery's quality control text and Linderman and colleagues in the Journal of Operations Management provide the sources in APA 7.

GM504 Unit 5 help: mistakes that cost points

GM504 Unit 5 plans often name a measure without defining it, so different people will count it differently. Others skip any test of whether the data can be trusted, which leaves later results open to doubt. A common mistake is reporting only an average baseline without showing variation or stability. Some writers choose a baseline period during an unusual season. Supporting measures are frequently unrelated to the cause found in the earlier analysis. Write each definition so two people would code the same case the same way, test agreement among the people collecting the data and revise until it is acceptable, collect ten or twelve weeks of baseline, plot it on a control chart, state whether it is stable, ground the goal in the root cause and say who owns each measure until the control plan takes over.

Get a GM504 Unit 5 example written to your instructions

Say what your Unit 5 plan must measure and which records your workplace or case already keeps, even messy ones. Include the assignment wording and rubric. The first plan costs nothing, matches that brief, defines each measure with its boundary cases, shows the baseline week by week, and reaches you, ready to adapt, in 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 GM504 papers and MS in Management samples

GM504 Unit 5 questions, answered

What is an operational definition, and why does it matter?

It is a definition precise enough that two people applying it to the same case reach the same answer. Without one, a later improvement might reflect a change in counting rather than a change in performance. The example spells out what counts as a correction and lists three cases that look like one but do not count.

Is an attribute agreement analysis required?

Rarely by name, but most rubrics reward some check on whether the data can be trusted. Where records are coded by people, having several coders classify the same sample is the simplest check available. For measured quantities such as time or weight, a gauge study plays the same role. A sentence acknowledging the limitation is the minimum.

How long should a baseline period be?

Long enough to show ordinary variation, which depends on how often the process runs. For a weekly rate, ten to twelve weeks is common; for a daily measure, a month may do. A baseline taken during an unusual period, such as a seasonal peak or a system changeover, should be flagged, since it will make any later comparison misleading.

Where can I find a free GM504 Unit 5 sample paper?

This page carries a full GM504 Unit 5 measurement and baseline plan for delivery corrections, with operational definitions, an attribute agreement test, a twelve-week baseline on a control chart and APA 7 references.

What is an attribute agreement test?

Several people independently classify the same set of records, and their judgments are compared. Here three billing clerks coded forty credit memos by reason; they agreed on 29 at first and on 37 after the definitions were clarified.