WMBA 6040 Week 3 Performance Measures Defense Example

Five Measures for an Optical Lab, Each Defined So Two Analysts Get the Same Number and Each Tied to What a Manager Does When It Moves: Coated Turnaround, Oldest Tray at Coating, First-Pass Yield, Frame Hold and Order Completeness

Student Name

Walden University

WMBA 6040: Improving Business Performance

Instructor Name

Month Day, Year

Five Measures for an Optical Lab, Each Defined So Two Analysts Get the Same Number and Each Tied to What a Manager Does When It Moves: Coated Turnaround, Oldest Tray at Coating, First-Pass Yield, Frame Hold and Order Completeness

Why So Few

The map drawn in Week 2 showed that a coated progressive job at Clearwater, the fictional eyeglass lab from Week 1, sits in the lab for roughly 8.3 working days yet receives less than three hours of actual work. The lab already reports dozens of figures, mostly counts of jobs shipped by department. None of them told managers where the 8.3 days went until the map was drawn.

This paper proposes five measures for the lab's production manager. Each must meet two tests. The definition must be tight enough that two analysts working separately from the same scan data arrive at the same figure. And each must be tied to an action a manager would take when it moves. Neely et al. (1995), reviewing research on the design of performance measurement systems, defined a performance measure as a metric that quantifies the efficiency or effectiveness of an action, and stressed that measures work as a designed system, not as isolated figures. Both points shape the proposal: each measure below is tied to an action, and the five are meant to be read together.

The Five Measures

Table 1

Measures, Definitions and Triggered Actions

Definitions use the lab's existing tray scans; thresholds are starting proposals to be reset after two months of data.

Why Each Definition Is Written as It Is

Coated turnaround starts at the order-release scan, not at order receipt. That choice separates delays inside the lab from delays caused by practices, which the frame hold and order completeness measures capture. Without the separation, a week in which many practices sent frames late would look like a production failure. The measure reports the 90th percentile as well as the median because practices remember the slowest jobs, and because averages can improve while the worst cases get worse.

The oldest tray at coating is a leading indicator of turnaround. Turnaround is known only when a job ships, days after the delay occurred. The oldest tray shows the problem while it can still be fixed. It is reported by formulation because the Week 2 map showed that coating delays come from waiting for a run of the right type. Hopp and Spearman (2008) emphasized that queue time at a constrained resource is usually the largest component of cycle time, which makes the queue at coating the point to watch.

First-pass yield counts only first inspections. If remade jobs were included, a week with many remakes would show a large number of inspections and could hide how many failed. Reason codes are required with every reject, because yield alone does not tell the quality lead what to fix.

Frame hold measures the share of jobs held more than three days rather than the average hold, because most frames arrive within a day and the problem is a small number of long holds. Order completeness is reported by practice, since the action, guidance and a call, is aimed at specific practices.

Measures Considered and Left Out

Jobs shipped per day is left out of this set. It measures volume, and it can rise while turnaround worsens, for example when the lab clears easy uncoated jobs first. Revenue per job is a finance measure that production cannot act on. Machine utilization in coating is left out because pushing utilization higher at a constrained resource tends to lengthen queues, the opposite of the goal; Little (1961) showed that, for a given rate of jobs, longer time in the system means more work in process, and the lab already has about 5,400 jobs in trays.

How the Measures Work Together

The five measures form a small chain. Order completeness and frame hold describe what arrives. The oldest tray at coating describes the constraint. First-pass yield describes the largest source of rework. Coated turnaround is the result the practices see. If turnaround worsens while the oldest tray stays low, the problem is elsewhere, most likely yield. If the oldest tray rises first, coating scheduling is the place to act.

Testing the Definitions

Before adoption, two analysts in the lab's office will calculate all five measures independently for the same past month from the raw scan file. Any measure on which they disagree will have its definition revised until they agree. This is the practical meaning of a definition tight enough to produce a single number.

Reporting and Review

The measures will be reported in one place: a single page posted each Monday in the production office and included in the weekly operations report to the general manager. The page will show each measure's current value, its value four weeks earlier and the trigger level beside it, so anyone glancing at it knows which actions are due that week. The production manager will review the page with the coating supervisor, the quality lead and the customer service manager in a fifteen-minute meeting every Monday morning. The meeting has one agenda: which triggers fired last week and whether the triggered action happened. Measures that never fire for three months will be reviewed to see whether the threshold is too loose.

References

Hopp, W. J., & Spearman, M. L. (2008). Factory physics (3rd ed.). McGraw-Hill/Irwin.

Little, J. D. C. (1961). A proof for the queuing formula: L = λW. Operations Research, 9(3), 383-387. https://doi.org/10.1287/opre.9.3.383

Neely, A., Gregory, M., & Platts, K. (1995). Performance measurement system design: A literature review and research agenda. International Journal of Operations & Production Management, 15(4), 80-116. https://doi.org/10.1108/01443579510083622