BUSI 3010 Week 10 Process Improvement Proposal Example

Stage the Next Slab and Check the Program While the CNC Is Still Cutting: An Improvement Proposal to Move 0.6 Hours of Setup Out of the Constraint's Cycle, Costed at $22,200 a Year, Owned by the CNC Lead and Measured by Hours per Job

Student Name

Walden University

BUSI 3010: Operations Management

Instructor Name

Month Day, Year

Stage the Next Slab and Check the Program While the CNC Is Still Cutting: An Improvement Proposal to Move 0.6 Hours of Setup Out of the Constraint's Cycle, Costed at $22,200 a Year, Owned by the CNC Lead and Measured by Hours per Job

The Problem as the Analysis Found It

The bottleneck analysis earlier in this course found that the edge-profiling CNC at the composite Fargo stone shop limits everything to 15.2 kitchen jobs a week. Its measured cycle time is 3.3 hours per job, against a standard program time of 2.5 hours. Spring sales of 17 to 18 jobs a week outran it, so the queue in front of the CNC grew and installations slipped from ten working days after templating to fourteen.

The 0.8 hours between standard and measured time is not cutting time. Timing showed where it goes: about 0.35 hours per job idle until the crane delivers the following slab section, about 0.25 hours verifying the program on the first piece after the slab is already on the table, and about 0.2 hours of slower feed on dark quartz, which the separate quality changes now in progress will address.

The Change

The proposal is to move the first two of those activities outside the CNC's cutting time. While the CNC is cutting one job, a material handler stages the next job's slab sections on the cart beside it and the CNC lead checks the next program on the second monitor against the slab photo. When the machine finishes, the next job is loaded immediately, and the program has already been verified.

This is the logic Shingo (1985) applied to machine changeovers: sort setup tasks into those that truly need the machine idle and those that do not, then shift the second group into the time the machine is busy. McIntosh et al. (2000) cautioned that the method's gains depend on redesigning the work around the machine rather than simply speeding it up, which is why this proposal adds a person and a monitor rather than asking the operator to hurry.

Expected Effect

Moving 0.35 hours of crane waiting and about 0.25 hours of program checking outside the cycle would bring the CNC's time per job from 3.3 hours to about 2.7 hours. At 50 scheduled hours a week, capacity would rise from 50 ÷ 3.3 = 15.2 jobs to 50 ÷ 2.7 = 18.5 jobs.

That covers the summer peak of about 18 jobs a week, which means the queue in front of the CNC should stop growing in May and June. If the dark quartz feed-rate changes also hold, the remaining 0.2 hours would shrink, and the CNC would approach its standard of 2.5 hours, or 20 jobs a week. At that point the CNC would still be the constraint, with installation the next limit at about 25 jobs a week.

Cost

Table 1

Annual Cost of the Change

Red River Stone & Surface and its costs are composite. Later years cost about $21,000.

The handler's hours are concentrated in the mornings and early afternoons, when the CNC runs most jobs. Outside those hours the handler sorts remnants in the yard, work currently done by fabricators on overtime.

Return

Each additional job the shop can complete is worth about $1,900 in contribution margin, the sale price of about $4,800 less stone, hardware and direct labor. Last May and June the shop lost 22 quoted jobs in which customers cited the lead time as their reason for going elsewhere. If the change lets the shop complete one additional job a week across the year, a conservative assumption given the summer losses, the contribution is about $95,000 a year against a cost of about $22,000. Even at half a job a week, the change pays for itself more than twice over.

Owner

The CNC lead owns the change. He will write the staging sequence with the handler, set the rule that no slab goes onto the CNC table until its program has been checked on the second monitor, and report the measures below each Monday. The production manager approves the handler's hiring and reviews the results.

How We Will Know It Worked

The primary measure is CNC hours per job, taken from the machine's run log each week. The target is 2.8 hours or less within eight weeks of the handler starting. Two supporting measures follow from it: jobs completed per week, which should rise above 17 in a week when demand is there, and the count of jobs parked beside the CNC at 3 p.m., which should fall from an average of 3.8 to below 2.

If CNC hours per job have not fallen below 3.0 after eight weeks, the change is not working as intended, and the handler's hours should be reassigned before the cost continues. Goldratt and Cox (2004) insisted that time lost at the constraint is lost to the whole plant, and that time saved at other stations buys nothing; this measure tests whether the hours saved are actually at the bottleneck.

What This Proposal Does Not Do

It does not add CNC hours, buy a second machine or move the CNC. Those are larger changes that may be needed if demand keeps growing, but they should wait until this change shows how much capacity the existing machine actually has. Spending on a second machine before recovering the first one's lost hours would buy capacity the shop already owns.

References

Goldratt, E. M., & Cox, J. (2004). The goal: A process of ongoing improvement (3rd rev. ed.). North River Press.

McIntosh, R. I., Culley, S. J., Mileham, A. R., & Owen, G. W. (2000). A critical evaluation of Shingo's 'SMED' (Single Minute Exchange of Die) methodology. International Journal of Production Research, 38(11), 2377-2395. https://doi.org/10.1080/00207540050031823

Shingo, S. (1985). A revolution in manufacturing: The SMED system. Productivity Press.