Predictika · Industrial boiler quotes

An AI sales agent's package boiler quote, checked before it goes out

An AI sales agent prepares a quote for an industrial water-tube package boiler: a model, a steam duty, a design pressure and temperature, a fuel and a few design options. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the published model ranges and design rules and returns a verdict, every reason and the figures it computed.

Scope: Industrial water-tube package boilers (FM, HCFM, PFM, PFI, PFT and TSSG families) from the public product brochure. See everything this validator covers

This validator models package boiler quotes, not just the scenarios shown below.

  • Duty and limits — the steam capacity range of each model, and the maximum design pressure and steam temperature published for the models that state them.
  • Fuel — package boilers burn liquid or gaseous fuels; solid fuels need a different boiler type.
  • Design options — a retention time of 5 minutes or more and separate drum shipping need the elevated-drum design, and lower-quality feedwater needs the MultiCirc option.

Any package boiler quote that touches these rules is in scope — not only the scenarios pictured here. Not covered: prices (boilers are engineered to order and this validator produces no price build-up), burner and emissions equipment, field-erected boilers, and pressure or temperature limits for the FM models, which the brochure does not state. Assumption: the 5 percent top-of-range flag is a sizing guideline set by this model, not a vendor rule.

The AI sales agentBuilds the boiler quote from the conversation and says it is ready.
Predictika validatorChecks every rule at once and returns the verdict, each finding, and the computed figures.

7 scenarios in 3 groups · each group runs from the agent’s first attempt to the outcome · click one to switch

What the agent proposes

The AI sales agent says

Illustrative agent statement. The validator receives the structured request below.

Proposing agent: AI sales agent (scripted)

Before sending, the agent must call Recorded run

Predictika validator


      
Waiting for the call.

Where rules-based CPQ gets it wrong

Click to expand

Our assessment of what a typical rules-based CPQ (product rules, price rules and validation rules, as in Salesforce CPQ) does with the same cases. It is a judgment about effort and reliability, not a claim that nothing can be built.

What the validator caughtTypical rules-based CPQPredictika
Capacity, pressure and temperature limits that differ by modelOnly with care
A product rule per model and per limit. It works, but every model is another set of rules to maintain.
The model and the duty travel in one request and every limit is looked up and compared on every call.
Design options that depend on other choices: elevated drum, MultiCirc, separate shippingOnly with workarounds
Each option links to a choice on another line, so it needs cross-line rules that interact with each other.
Every option and the choices it depends on are in one request and checked together.
Retention time that forces a designOnly with workarounds
A threshold on one field decides what another field must be, and a missing companion option is usually caught late or not at all.
The retention time and the design option are checked together on every call.
Fuel that the product family cannot burnOnly with care
A fuel list per product, which has to be kept in step with the catalog.
The fuel is checked against the product family on every call.
Every problem returned in one answerOnly with workarounds
Rules tend to fire as the quote is saved or a step is completed, so a rep often fixes one error and meets the next.
All findings come back together, whatever order the choices arrived in.

What this validator catches: findings from its test cases

Click to expand

Every rule is checked on every call, in any order the choices arrive; one problem never hides another. Blocking findings stop the quote; approval findings need a person to sign off; everything else is computed.

1The required 5,000 lb/hr is below the 8,200 lb/hr minimum of this model.
2The required 300,000 lb/hr is above the 275,000 lb/hr maximum of this model.
3The required 300,000 lb/hr is below the 350,000 lb/hr minimum of this model.
4The required 650,000 lb/hr is above the 600,000 lb/hr maximum of this model.
5The required 580,000 lb/hr is within 5 percent of the 600,000 lb/hr maximum of this model; a person should confirm the sizing.
6The required 600,000 lb/hr is within 5 percent of the 600,000 lb/hr maximum of this model; a person should confirm the sizing.
7The design pressure of 1,300 psig is above the 1,250 psig limit of this model.
8The steam temperature of 850 F is above the 825 F limit of this model.
9The design pressure of 1,900 psig is above the 1,800 psig limit of this model.
10The design pressure of 1,200 psig is above the 1,150 psig limit of this model.
11The steam temperature of 1,010 F is above the 1,000 F limit of this model.
12The design pressure of 2,500 psig is above the 2,400 psig limit of this model.
13Industrial package boilers burn liquid or gaseous fuels; solid fuels such as coal, wood and biomass need a Stirling or Towerpak boiler.
14A retention time of 5 minutes needs the elevated-drum design, which is for retention times of 5 minutes or more.
15Lower-quality feedwater needs the MultiCirc design option, which is not on the quote.
16Separate shipping of the drum is a feature of the elevated-drum design, which is not on the quote.

Scenarios and results

Click to expand

The recorded result of the validator for each scenario on this page: the request the AI agent sent, the verdict, and every finding, in the order they appear above.

1. Model, pressure and temperature limits: HCFM boiler for 380,000 lb/hr at 1,400 psig and 830 F with poor feedwater

The agent's request. Please quote a B&W HCFM package boiler for 380,000 lb/hr of steam at 1,400 psig and 830 F, burning gas or oil, with a 3-minute retention time, on lower-quality feedwater. It is ready to send.

Verdict: Blocked.

  • Capacity above the model range. The required 380,000 lb/hr is above the 350,000 lb/hr maximum of this model.
  • Pressure above the model limit. The design pressure of 1,400 psig is above the 1,250 psig limit of this model.
  • Steam temperature above the model limit. The steam temperature of 830 F is above the 825 F limit of this model.
  • Lower-quality feedwater needs MultiCirc. Lower-quality feedwater needs the MultiCirc design option, which is not on the quote.

2. Model, pressure and temperature limits: The agent resubmits after fixing the errors

The agent's request. Please quote a B&W PFM package boiler for 380,000 lb/hr of steam at 1,100 psig and 800 F, burning gas or oil, with a 3-minute retention time, MultiCirc, on lower-quality feedwater. It is ready to send.

Verdict: Allowed.

  • Order summary. 380,000 lb/hr of steam at 1,100 psig and 800 F.

3. Fuel and design options: Package boiler quote for solid fuel with long retention and separate shipping

The agent's request. Please quote a B&W PFM package boiler for 400,000 lb/hr of steam at 1,400 psig and 830 F, burning solid fuel, with a 6-minute retention time, separate drum shipping. It is ready to send.

Verdict: Blocked.

  • Solid fuel on a package boiler. Industrial package boilers burn liquid or gaseous fuels; solid fuels such as coal, wood and biomass need a Stirling or Towerpak boiler.
  • Long retention time needs the elevated drum. A retention time of 6 minutes needs the elevated-drum design, which is for retention times of 5 minutes or more.
  • Separate shipping needs the elevated drum. Separate shipping of the drum is a feature of the elevated-drum design, which is not on the quote.

4. Fuel and design options: The agent resubmits after fixing the errors

The agent's request. Please quote a B&W PFM package boiler for 580,000 lb/hr of steam at 1,400 psig and 830 F, burning gas or oil, with a 6-minute retention time, the elevated drum and separate drum shipping. It is ready to send.

Verdict: Needs approval.

  • Capacity near the top of the model range. The required 580,000 lb/hr is within 5 percent of the 600,000 lb/hr maximum of this model; a person should confirm the sizing.
  • Order summary. 580,000 lb/hr of steam at 1,400 psig and 830 F.

5. Fuel and design options: The agent resubmits after fixing the errors

The agent's request. Please quote a B&W PFM package boiler for 500,000 lb/hr of steam at 1,400 psig and 830 F, burning gas or oil, with a 6-minute retention time, the elevated drum and separate drum shipping. It is ready to send.

Verdict: Allowed.

  • Order summary. 500,000 lb/hr of steam at 1,400 psig and 830 F.

6. Capacity range: PFI boiler quote above its capacity range

The agent's request. Please quote a B&W PFI package boiler for 720,000 lb/hr of steam at 1,100 psig and 900 F, burning gas or oil, with a 3-minute retention time. It is ready to send.

Verdict: Blocked.

  • Capacity above the model range. The required 720,000 lb/hr is above the 700,000 lb/hr maximum of this model.

7. Capacity range: The agent resubmits after fixing the errors

The agent's request. Please quote a B&W PFI package boiler for 650,000 lb/hr of steam at 1,100 psig and 900 F, burning gas or oil, with a 3-minute retention time. It is ready to send.

Verdict: Allowed.

  • Order summary. 650,000 lb/hr of steam at 1,100 psig and 900 F.

Narration transcript

Click to expand

The text of the narrated walkthrough, step by step.

  1. Welcome to a demo of Predictika's Validator. An AI sales agent prepares a quote for an industrial water-tube package boiler: a model, a steam duty, a design pressure and temperature, and more. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the published model ranges and design rules and returns a verdict, every reason and the figures it computed.
  2. I will start with the first request by the agent.
  3. The agent says. Please quote a B&W HCFM package boiler for 380,000 lb/hr of steam at 1,400 psig and 830 F, burning gas or oil, with a 3-minute retention time, on lower-quality feedwater. It is ready to send.
  4. Before it goes out, the agent calls the validator.
  5. Blocked, with four findings, all in one answer. The required 380,000 lb/hr is above the 350,000 lb/hr maximum of this model. The design pressure of 1,400 psig is above the 1,250 psig limit of this model. The steam temperature of 830 F is above the 825 F limit of this model. The other one is on screen.
  6. The validator returned four blocking findings in one answer. A check that reads one field at a time would pass most of this; it shows only when the choices in the quote are checked against each other.
  7. What this shows: for capacity, pressure and temperature limits that differ by model, a rules-based CPQ is error-prone for this.
  8. Now the agent resubmits.
  9. The corrected request. It changes the model to PFM, the design pressure to 1,100 psig, the steam temperature to 800 F and the MultiCirc to yes.
  10. It calls the validator again.
  11. This time the validator clears it.
  12. All four earlier findings are cleared.
  13. Four problems caught in one call, fixed in one pass, and a clean quote. That is the value of Predictika's validator: catching mistakes before they reach the customer. Now you can try the other scenarios yourself.