I85PREP
⌘ K
CSSYB study materials
CSSYB study resources

An ERP-Shaped Charter at a Molding Plant

CSSYB · Put your knowledge to work in context.

All case studies

An ERP-Shaped Charter at a Molding Plant

At a 60-person injection-molding plant, the operations director has chartered a project: "Implement the new ERP scheduling module to improve on-time delivery and reduce defects to 99% quality; scope: the entire order-to-cash process across all three plants; complete by Q4." The team has eight members, two of them consultants from the ERP vendor. Customer scorecards show on-time delivery averaged 88% over the last six months against a 98% requirement, and one large customer has issued a chargeback notice. There is no baseline dataset, no financial validation, and the first team meeting is next week. The internal defect rate on the main part family is 3.1%.

1. Rewrite the problem statement and goal statement. Identify each defect in the original charter and the fix applied.

Read the worked answer

Defects in the original:

  1. Solution embedded - "implement the ERP module" is a countermeasure, not a problem.
  2. No baseline - nothing quantified from data.
  3. Arbitrary target - "99% quality" names no measure (defect rate? yield? DPMO?).
  4. Scope too large - three plants, order-to-cash, one quarter.
  5. No customer linkage - the CTQ (98% OTD) is not in the statement.
  6. Deadline from the calendar, not from capability.

Rewritten problem statement: "From January through June, on-time delivery for the [part family] shipped from Plant 1 averaged 88% against a customer requirement of 98%, generating $__ in chargebacks and expedite cost (source: customer scorecards and freight invoices)."

Rewritten goal statement: "Raise on-time delivery for that part family to at least 98% by [date] with no increase in freight cost, verified by three consecutive months of scorecard data."

Move the ERP decision to a parking lot / potential countermeasure. If the capital commitment is already made, list it as a project constraint so the team does not fight a settled decision.

2. The customer scorecard also penalizes "short shipments" and "packaging damage." Build a CTQ tree and write an operational definition for one CTQ.

Read the worked answer

CTQ tree:

  • Need: my order arrives complete and on time, undamaged.
    • CTQ 1: On-time delivery
    • CTQ 2: Order completeness (no short shipments)
    • CTQ 3: Packaging integrity

Operational definition (Order completeness): "An order is complete if the quantity shipped on the packing list equals the quantity ordered on the customer PO at the moment of shipment. Short shipments caused by customer-requested partial releases are excluded and coded separately."

For on-time delivery the definition must pin down: which date is the target (customer requested date, not promise date), which date proves delivery (carrier proof of delivery vs. ship date), and whether early counts as on time.

Why this matters for this project: the internal 3.1% defect rate is not the customer's measure. Before any baseline is credible, the team must confirm that the metric on the scorecard can be reproduced internally from shipment and POD data. If it cannot, that gap is itself a Define-phase finding and a data-collection risk.

Finally, note the definitions must be signed off by the customer-facing owner, not invented by the team.

3. The sponsor insists all three plants stay in scope. How do you handle it?

Read the worked answer

Do not simply comply, and do not refuse - renegotiate with evidence.

  1. Quantify the cost of the scope. Stratify chargebacks and late orders by plant. If one plant drives most of the COPQ, that is the defensible pilot.
  2. Show the arithmetic of full scope. Three plants means stratified data collection by plant, line, shift and customer, plus separate MSAs - realistically a year, not a quarter. Put that timeline in front of the sponsor alongside the Q4 date he already committed to.
  3. Offer a structure that protects his goal: one plant/one part family in scope now; the other two plants entered as documented replication after the first shows a result. He gets the Q4 deliverable; the other plants get a proven package.
  4. Make the risk his to weigh, in his language: "full scope, no result by Q4" vs. "one plant, measured result by Q4, roll-out after."

Whatever is agreed, write it into the charter with the sponsor's signature. Scope that is not recorded will be re-expanded the first time a meeting gets uncomfortable.