In short
An engineering change order (ECO) changes a design after it was meant to be stable. Its cost is rarely in the change itself; it is in the downstream work it invalidates — timing, power, area, design-rule checks and verification. The impact should be modelled at the moment the change is proposed, before the change board approves it.
Why are late ECOs dangerous?
Late in physical design, the design is finely balanced. A small logic change can break timing closure in a block that took weeks to close, or push an integration regression into another loop.
What should the change board see?
Alongside the change itself:
- Which blocks and milestones it touches.
- The expected effect on timing, power and area.
- What has to be re-verified.
- The effect on the date, with a range.
Who approves?
The same approvers as today. The change board is a ritual worth keeping. What changes is that the impact is in front of them when they decide, instead of being discovered afterwards.