CONTEXT(CLONE)
Sign in

Reconcile study assumptions across an interconnection queue

For: Interconnection manager at a developer, or a transmission planner reviewing clustered studies

Pattern: Map, verify, reduceNeeds scaleDesigned for 30 to 400 agents
Get early accessSign in to use

The pain today

Each project's study rests on a base case, on earlier-queued projects assumed in service, on short-circuit levels and on shared network upgrades. Across a queue these drift apart, and an upstream withdrawal quietly shifts cost allocation.

The ask

“I attached the impact and facilities study reports and the queue list for our projects and the ones ahead of us. For each project, pull out the base case, which earlier projects are assumed in service, the point of interconnection, reactive power and short-circuit assumptions and the allocated upgrades. Show me where two studies assume different things about the same substation, line or upgrade.”

Plain words, as you would say it to a colleague. Edit it to fit your case before you send it.

What you attach or connect

  • System impact and facilities study reports as text
  • Queue list with status and withdrawal notes
  • Interconnection agreements and cost allocation tables
  • Study scope and procedure documents

The unit of work

One worker task per one project's study report.

Why a swarm fits

Each report is read alone against a fixed list of assumptions. The value is in the comparison: the same transformer, line or upgrade described differently in reports nobody reads side by side.

Not for

It reads what the reports state. It does not run load flow, short-circuit or stability studies and cannot say which assumption is electrically right.

The decision tree

6 typed decisions, each with an action for every answer

At fixed moments in a run, the engine puts one narrow question to a decision model. The decision model never writes text: it answers yes or no with a probability, picks from listed options, or gives a score, about a small slice of the material. The engine then does exactly what this tree says, which is what makes the run auditable. The thresholds are the template's design values, not measured results.

  1. Planner, while planning

    Scope checkYes or no, with a probability

    Before work starts on a unit

    Is this document a study report for a project on the supplied queue list, in its latest listed revision?

    Sees only: The report's title page, revision block and the queue list row

    Why: Keeps superseded revisions and unrelated studies out of the comparison.

    • Yes: 0.60 or higherthenAccept
    • Unsure: 0.30 up to 0.60thenEscalate to a strong model
    • No: below 0.30thenSkip this unit

  2. Before workers, before a task runs

    Small worker or strong modelA choice among options

    Before a task runs

    Does this report state its base case, in-service assumptions and upgrades in tables, or only in running text and appendices?

    Sees only: The report's table of contents and assumptions section headings

    Why: Reserves the strong model for reports whose assumptions have to be pieced together.

    • Stated in tablesthenAccept
    • Spread through prose and appendicesthenEscalate to a strong model

  3. After workers, the judge checks

    Evidence checkA choice among options

    After a worker answers

    Does the quoted passage state the short-circuit ratio or fault level assumed at this project's point of interconnection, rather than a value for a different bus?

    Sees only: One extracted value, the quoted passage and the project's point of interconnection name

    Why: Stops a neighbouring bus value from being recorded as this project's assumption.

    • States it for this point of interconnectionthenAccept
    • States a value for a different busthenReject and retry
    • Bus not named in the passagethenMark unresolved

  4. Evidence checkYes or no, with a probability

    After a worker answers

    Does the quoted passage name this earlier-queued project as in service or included in the base case?

    Sees only: One queue project name and the quoted passage from the later study

    Why: Makes every dependency on an earlier project traceable before a withdrawal is propagated.

    • Yes: 0.85 or higherthenAccept
    • Unsure: 0.50 up to 0.85thenEscalate to a strong model
    • No: below 0.50thenReject and retry

  5. Reconciler, while merging

    Conflict checkA choice among options

    While reconciling

    Do these two passages from different studies refer to the same substation, line or network upgrade?

    Sees only: Two quoted passages and the queue list's facility names

    Why: Finds conflicting assumptions about shared facilities while refusing to guess at facility identity.

    • Same facility, same assumptionthenAccept
    • Same facility, different assumptionthenMark unresolved
    • Different facilitiesthenContinue
    • Naming too ambiguous to tellthenAsk a person

  6. Accountable person, before anything is settled

    Person decidesYes or no, with a probability

    Before anything is reported as settled

    Does the finding concern allocated upgrade cost, a withdrawn project still assumed in service, or an assumption at our own point of interconnection?

    Sees only: One reconciled finding with its quotes

    Why: Sends anything bearing on study validity or restudy to the licensed engineer and interconnection manager.

    Accountable: A licensed power systems engineer owns any conclusion about study validity; the interconnection manager owns restudy requests.

    • Yes: 0.40 or higherthenAsk a person
    • Unsure: 0.15 up to 0.40thenAsk a person
    • No: below 0.15thenAccept

The fleet: who does what

Model tiers by role, not brands: you choose the models. Strong reasoning models plan and reconcile, small fast models do the wide work, and the judge is a decision model from a different family, so it does not share the workers' blind spots.

  1. Planner

    A strong reasoning model fixes the assumption schema and a common naming for buses, lines and upgrades across reports.

    Decisions here:1. Scope check

  2. Workers

    Small fast workers from an open-weight family extract one report's assumptions with page-level quotes.

    Designed for 30 to 400 agents, one worker task per one project's study report. Each worker receives only its own unit.

    Decisions here:2. Small worker or strong model

  3. Judge, from a different model family

    A decision model from a different family checks that each extracted value is stated in the quote, not inferred.

    Decisions here:3. Evidence check4. Evidence check

  4. Reconciler

    A strong reasoning model lines projects up per shared facility and lists conflicting assumptions and stale dependencies.

    Decisions here:5. Conflict check

  5. Accountable person

    A licensed power systems engineer owns any conclusion about study validity; the interconnection manager owns restudy requests.

    Decisions here:6. Person decides

Checked before anything is accepted

  • Every assumption carries a quote and the report revision it came from
  • Facility names are matched through the supplied queue list, and uncertain matches are flagged
  • Projects marked withdrawn are traced to every study that still assumes them

What comes back

  • Assumption table per project with sources
  • Conflicts per shared substation, line or upgrade
  • Studies that depend on withdrawn or changed projects
  • Questions to raise with the system operator

What to measure

  • Conflicts confirmed by the planning engineer
  • Assumptions extracted without a supporting quote
  • Engineer hours per study reviewed
  • Cost per report

Names of measures only. No result is claimed for this template.

Templates open in the workspace chat with the ask filled in. Nothing runs until you send it.

Get early accessSign in to use

More in Energy and utilities

Check a flexibility contract portfolio against dispatch obligations

For: Portfolio manager at a flexibility aggregator, or a utility demand-response programme lead

Every site contract has its own notice time, duration limit, activation cap, availability window, baseline method and penalty.

Pattern: Specialist panelNeeds live models6 decisionsDesigned for 8 to 300 agents

Review maintenance and test records across a substation fleet

For: Asset manager or maintenance engineer at a transmission or distribution network operator

Oil analysis reports, breaker timing tests, relay test sheets and battery discharge tests sit inside work orders.

Pattern: Hierarchical decompositionNeeds a connector6 decisionsDesigned for 40 to 600 agents

Trace grid code and tariff changes to every affected asset

For: Grid compliance or regulatory affairs manager at a generator, storage operator or network company

A revised grid code clause or tariff schedule touches some assets and not others, depending on connection date, voltage level, technology and derogations.

Pattern: WatchtowerNeeds a connector6 decisionsDesigned for 20 to 500 agents