Baseline gates
Six phase bands structure the baseline. Project instances vary from 60 to 72 rows with conditional scope.
Requirements engineering / applied research
Make obligations traceable from the first requirement to the evidence that closes them.
Deployed instrument · Preliminary, single-organization observations
Line-item first-pass approval
69%
20 first-pass approvals / 29 dispositioned line items
Different units. Not a performance trend.
01 / the engineering problem
A small subcontractor must track preparation, prime routing, government review, evidence, deadlines and approvals that can expire. The public model documents a deployed instrument in one organization, while keeping operational records private.
Six phase bands structure the baseline. Project instances vary from 60 to 72 rows with conditional scope.
Preparation, routing, adjudication, closure and expiry each retain meaning and evidence guards.
Preliminary observations from one organization. No control group or causal effect established.
02 / traceability in motion
Simplified explanatory path through the 11-state model. Rejection and renewal loops remain in the source specification; this is not a live instrument view.
03 / explore a dependency
A site-dependent submittal can be due before its prerequisite is available. Move the access day to see the timing relationship. This synthetic example assumes a deadline of award + 30 calendar days and zero preparation time after access.
A negative margin identifies a conflict under those assumptions. A nonnegative margin does not establish practical feasibility, contractual compliance or an approved remedy.
Timing conflict: access arrives 10 days after the due date.
Synthetic illustration · calendar days · not operational data
04 / measurement discipline
The 11 September 2026 snapshot reports 20/29 first-pass approvals by dispositioned line item and 8/17 by package. A package can contain several line items. The switcher above changes the analytical unit, not the time period.
The dashboard's 85% current-state gate share answers a different question. Superseded and earlier gate-row percentages are not combined into an improvement claim.
Sanitized architecture and bounded aggregate observations. Raw operational data are not public; this review did not independently reproduce the measurements. No population-wide result, causal improvement or established scholarly novelty is claimed.
05 / authorship and next work
Independent work by Jason V. Holmes, published under CC BY 4.0. No employer, university or government endorsement is claimed. Literature review comes before expanded taxonomy or stronger novelty claims; independent coding and a broader comparative design remain research work.
Systems, evidence, iteration
I connect requirements to models, interfaces and evidence, using AI-assisted implementation under human review.
Start a conversationPortfolio review: 17 November 2026.