Service RD5A
Model Development
Runs transactional. Starts when a scheduled window opens. It runs 2,326 times a week across 8 stations. 2 of those stations stop for a person and 6 are done by an agent, 4 of them on the shared spine.
Runs a week
2,326
94 in flight right now.
Stations
8
2 person · 2 agent · 4 spine
Agent share of steps
75%
Autonomy on this service is set at 75% against a ceiling of 83%.
Decisions a person makes
2
Held by Director, Data Science.
Touchless
76%
Share of runs that complete without anybody opening them.
Timetable against actual
16.1h
Actual 13.6h, 2.5h under the timetable.
Operations
The lane diagram
Read it left to right. more
Operations
Every station in order
The same stations as a list, with the dwell the estate records against each one and the system a station of that shape would touch.
| # | Station | Lane | Who | Layer | Dwell | Target | In flight | System |
|---|---|---|---|---|---|---|---|---|
| 1 | Frame problem | Agent | Feature Engineering Agent | workflow | 44m | 1h | 0 | Product Lifecycle Management |
| 2 | Engineer features | GBS | Feature Engineering Agent | gbs | 2h 32m | 4h 31m | 31 | Product Lifecycle Management |
| 3 | Train | GBS | Feature Engineering Agent | gbs | 1h 29m | 3h 30m | 13 | Product Lifecycle Management |
| 4 | Validate | GBS | Feature Engineering Agent | gbs | 2h 59m | 3h 57m | 27 | Product Lifecycle Management |
| 5 | Explainability pack | Agent | Explainability Agent | workflow | 1h 23m | 2h 27m | 4 | Research Notebook and Experiment Data |
| 6 | Reviewer approve Held for a named human when the unit exceeds the model decision envelope, when confidence falls below the floor, or when a policy clause is engaged. | Human | Director, Data Science | policy | 1h 21m | 2h 24m | 9 | Product Lifecycle Management |
| 7 | Register | GBS | Feature Engineering Agent | gbs | 2h 10m | 3h 1m | 2 | Product Lifecycle Management |
| 8 | Deploy Held for a named human when the unit exceeds the model decision envelope, when confidence falls below the floor, or when a policy clause is engaged. | Human | Director, Data Science | policy | 3h 1m | 6h 50m | 8 | Product Lifecycle Management |
Operations
Run it
Change the volume, move the autonomy line, release a gate, and watch what happens to the hours in each lane and to the time it takes one run to get through. more
The simulation is arithmetic on the estate, not a queueing model. more
How it runs
Run it again, going wrong
The same unit and the same stations, broken the way the estate already records breaking. more
8 stations, walked twice
Once clean, once with the categories the estate counts
This is the same unit on the same stations, breaking. more
No break below was chosen. more
One play head, two tracks, the same scale. The clean run stops when it runs out of stations; the other one keeps going.
As written
8 stations, nothing waiting, nothing sent back
0 min of 16 hrs
now Frame problem · Feature Engineering Agent · 44 min
As the ledger has it
the same stations, plus 6 categories the estate already counts
0 min of 2.1 days
now Frame problem · Feature Engineering Agent · 44 min
What breaks, and where it is counted
No break below was chosen. more
What the two runs say together
6 counted categories land on this procedure. more
What this run cannot claim
These breaks are shown together on one unit so the shape of a bad day fits on one screen. more
The separation register records a duty conflict against a sub-function and a named post, not against one procedure. more
Two clocks run. Worked time is the sum of the time the work is actually being done … more
Live observability
Systems this service would touch
Attributed by rule from the function's system list. more
2 systems attributed
0 of 2 returning records
Product Lifecycle Management
Siemens Teamcenter · engineering · inbound · high
7 of 8 stations attributed here · read only
Research Notebook and Experiment Data
Benchling · engineering · inbound · medium
1 of 8 stations attributed here · read only
System attribution is a rule, not a reading. more
Who runs it
Orchestration and accountability
- Orchestrated by
- Modeling Orchestrator
- That orchestrator’s remit
- Owns model work end to end inside Research & Development. Runs 3 lines across 19 stations, drains the repeatable share into the DATA-ANALYTICS tower, and holds 4 human gates.
- Accountable person
- Director, Data Science
- Sub-function
- Data Science & Modeling
- Shared spine tower
- data-analytics
- Hand-over points
- None — this service stays inside its function
Other services in this sub-function
2 alongside this one
What this page is and is not
This is a catalogue of the modeled estate. more