What just happened
An IP vendor emails one engineer: the SerDes delivery will be four weeks late. Six blocks use it.
Why it is hard
The delay does not hit every block the same way, the possible fixes compete for the same timing engineers, and every day spent working out the answer removes an option.
Where Tymeline works
- Supplier email
- Timing reports
- Jira
- Slack
- TraceFollows the delay through every block that uses the IP. Three absorb it. The coherent fabric cannot: it moves 28 days, and samples move 11.
- SimulateModels three recoveries against the real plan: re-sequence fabric timing with a parallel run, switch to an alternate IP drop, or cut the fabric feature. Only the first recovers all 11 days without added cost or lost scope.
- PrepareBuilds the decision packet: the new sequence, who moves, what it asks of each team, and the escalation to the vendor.
- ExecuteOn approval, re-sequences the work in Jira, tells every owner whose dates change, sends the vendor escalation and signs the record.
You decideThe VP of Engineering approves the recovery, with MFA.