Case · a Western European rail electronics supplier

A bid managed with these tools, inside a client's team

This was not a demonstration. A rail electronics supplier engaged us as bid manager for a public tender, and the bid was run with the systems described on this site — the Tender Wikipedia over the supplier's own material and the agents of the Bid Stack — from the day the documents were published to submission. What follows is what that produced, in the order it happened.

Within 24 hours: the bid overview

Key points, red flags, risks, contradictions between documents, and the questions worth putting to the tendering authority, read from several hundred pages in the authority's language. Work that conventionally takes a bid team its first two weeks.

Bid overviewLight-rail onboard signalling retrofit — 14 vehicles
day 1
Tender documents
  • ITT Vol. 2Technical requirements
  • Annex CAvailability
  • Price schedulePrice items
several hundred pages · German
Key points
Series retrofit of all 14 vehicles within 26 months of contract award.ITT Vol. 2 · §4.3.1
Red flags
Fleet availability is defined twice, differently.Annex C · §2 · ITT Vol. 2 · §6.2
Risks
One of three penalty regimes is unclear about whether a cap applies.ITT Vol. 2 · §4.3 · Annex C · §2
Contradictions
Software maintenance: five years in the requirements, three in the price schedule.ITT Vol. 2 · §8.3 · Price schedule · item 5.1
Questions for the authority
How is fleet availability measured while vehicles are out for retrofit?Annex C · §2
Every item cited to its source, read in the original language.

Fictional tender and findings, on the same retrofit as the other examples.

Back to timeline ↑

Within 24 hours: the penalty calculator

Three distinct penalty regimes in the contract, one of them unclear about whether a cap applied. We built an interactive calculator so the bid team and the managing director could see total exposure change as assumptions changed, rather than reading clauses.

Penalty exposureLight-rail onboard signalling retrofit — 14 vehicles
contract € 18.5 M
Regime · sourceYour planExposure
First vehicle back in service · € 30 k per week beyond month 14R-042 · ITT Vol. 2 · §4.3 · cap 5 %month 15€ 130 k
Series retrofit completed · € 45 k per week beyond month 26R-043 · ITT Vol. 2 · §4.3.1 · cap 7.5 %month 26€ 0
Fleet availability below 98.5 % · € 8 k per 0.1 pp per monthR-044 · Annex C · §2 · cap 4 % · definition open98.3 %€ 416 k
Total exposure at this planoverall cap 10 % · € 1.85 M
€ 546 k

Illustrative regimes and rates on the same fictional tender. Adjust the plan to see the exposure move.

Back to timeline ↑

Within the first week: the spare-parts calculator

A calculator derived from more than five hundred pages of technical documentation, produced in about 45 minutes including two review iterations, against an estimated week of a procurement engineer's time.

Spare-parts calculatorLight-rail onboard signalling retrofit — 14 vehicles
500+ pages
With the calculatorabout 45 min
Conventional estimateabout a week

including two review iterations · against a procurement engineer's time

PartSourcePer vehicleFleet · 14Spares
Onboard computer unitTech. doc. vol. 1 · p. 34Tech. doc. vol. 1 · p. 341142per vehicle 1 · fleet 142 spares
Driver displayTech. doc. vol. 1 · p. 118Tech. doc. vol. 1 · p. 1182283per vehicle 2 · fleet 283 spares
Balise antennaTech. doc. vol. 2 · p. 212Tech. doc. vol. 2 · p. 2122284per vehicle 2 · fleet 284 spares
Odometry sensorTech. doc. vol. 2 · p. 257Tech. doc. vol. 2 · p. 2572284per vehicle 2 · fleet 284 spares
Every quantity traced to the page it came from.

Fictional parts and quantities, on the same retrofit as the other examples.

Back to timeline ↑

Within the first week: the scope and cost map

Every scope element the tender required, laid against where a price item had been allocated in the bid documents or the Q&A. The gaps were visible at a glance, and were raised with the authority before they became a pricing problem.

Scope and cost mapLight-rail onboard signalling retrofit — 14 vehicles
2 gaps · 1 partial
IDScope elementsourcePrice itemStatus
S-01Onboard units, supply for 14 vehiclesITT Vol. 2 · §4.1Price schedule · item 1.1allocated
S-02Installation and commissioning, per vehicleITT Vol. 2 · §4.3Price schedule · item 2.1allocated
S-03Safety case update and type approvalITT Vol. 2 · §5.2Price schedule · item 3.4allocated
S-04Driver and maintainer trainingITT Vol. 2 · §7.1Q&A no. 17 — included in item 2.1via Q&A
S-05Spare parts for the retrofit periodITT Vol. 2 · §4.6none foundgap
S-06Measures to hold fleet availability during the retrofitAnnex C · §2none foundgap
S-07Software maintenance, five years after acceptanceITT Vol. 2 · §8.3Price schedule · item 5.1 — three years onlypartial
Gaps and the partial item raised as clarification questions before pricing.

Fictional tender, scope and price schedule, on the same retrofit as the other examples.

Back to timeline ↑

Throughout: source-verified outputs

Every claim in every document was checked back against the tender; adjacent text re-read to confirm the interpretation; assessments labelled as assessments. This is the reason the team could act on the analysis instead of re-doing it.

Extracted claimfact · verbatim

Series retrofit of the onboard signalling equipment shall be completed for all 14 vehicles within 26 months of contract award.

ITT Vol. 2 · §4.3.1 · p. 87 · R-043

Adjacent text, re-read in the original

„Das erste nachgerüstete Fahrzeug ist spätestens im 14. Monat nach Auftragsvergabe wieder in Betrieb zu nehmen; die Serien-Nachrüstung erfolgt in Losen von jeweils zwei Fahrzeugen.“

Assessmentassessment · not verbatim

Two vehicles out of service at any time between month 14 and month 26. Read the availability clause in Annex C §2 against this.

One requirement from a tender, as the verification layer returns it: the claim quoted, its source cited, the surrounding text re-read in German, and the assessment kept apart from the fact.

Back to timeline ↑

For the writers: response templates

Per-document templates referencing the exact tender sections, requirements and scoring criteria each answer had to hit, so the people writing could write rather than search.

Response templateLight-rail onboard signalling retrofit — 14 vehicles
3 requirements · 1 criterion3 req. · 1 criterion

Answer 4.3 · Delivery programme

responds to ITT Vol. 2 · §4.3
1 · Programme and batchesR-043 · ITT Vol. 2 · §4.3.1
2 · First vehicle back in serviceR-042 · ITT Vol. 2 · §4.3
3 · Holding fleet availabilityR-044 · Annex C · §2

Fictional template, on the same retrofit as the other examples.

Back to timeline ↑

What the team did with it

The systems took the reading and the cross-checking off the team's desks. The time that freed went into the work only the team could do.

Each specialist started from a sourced summary of the requirements that touched their area, rather than from the raw tender, and spent their hours on the answers, where their knowledge of the product and the customer made the difference.

The team as a whole met regularly. The bid manager kept everyone informed as findings and the authority's answers came in, and brought in legal, procurement, project management and finance wherever their expertise was needed. Their knowledge of the domain and of how the company works steered how the systems were used: which findings to pursue, which to set aside, and which questions to put to the authority. Because every point on the table carried its source, discussions moved quickly from what the tender said to what the supplier should do about it.

Where the systems fell short, it was in interpretation: a passage translated slightly out of context, or a requirement read more rigidly than it was written. The verification layer caught these before they reached a decision, and none led to a problem in the bid.

What this shows

The systems did the reading, the cross-checking and the drafting. The bid team did the judging. That division of labour is what we mean by AI in production, and it is available to any supplier with a tender on the desk.