Predictika · Chiller plant quotes

An AI sales agent's chiller plant quote, checked before it goes out

An AI sales agent drafts a quote for an air-cooled chiller plant: capacity at design conditions, sound, electrical load, options, lead time and price for one job site. Before the quote goes out, the agent sends it to Predictika, which checks it against a declarative model, compares the total the agent states with the one it computes, and returns every finding together with the line-item quote.

Scope: air-cooled chiller plant quotes: N identical chillers for one job site, priced as a line-item quote in dollars. See everything this validator covers

This is a CPQ validator for one thing: air-cooled chiller plants built from one chiller model per quote. It does not cover other HVAC equipment, and it is not an engineering design.

  • Site conditions — freeze protection or glycol for cold sites, high-ambient kits above 105F design ambient, seismic certification kits, and coastal corrosion review for uncoated coils.
  • Capacity and fit — delivered tonnage against the load at the design ambient and fluid, with or without a standby unit, roof weight, sound at the property line, and the 6 chiller limit per plant controller.
  • Electrical — model availability at the site voltage, minimum circuit ampacity against the electrical service, and a review when the plant is more than 1.5 times the load.
  • Schedule and price — lead time against the weeks the job allows with factory expedite approval, a line-item price roll-up, discount authority by equipment value, and the total the agent states.

Any plant built from these chiller models, sites, fluids, options and discounts runs through this validator, not just the scenarios shown on this page. It is an illustrative model; capacities, amps, weights, sound levels, prices and lead times are fictional estimates.

The AI sales agentBuilds the chiller plant 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.

8 scenarios in 4 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
Plant capacity at design conditions: each chiller derated for the site design ambient and the fluid, then the plant sized with one unit on standby against the loadOnly with workarounds
Needs chained formula fields for the derating, the running count and the plant total; each site or fluid change means checking every dependent field.		Derating, running units and plant capacity are derived on each call from the model and site data, and compared with the load together with every other rule.
Sound at the property line: unit count stacking, the low-sound option and the site attenuation against the site limitOnly with care
One rule per unit count and option combination against each site limit; a new site means more rules to write and keep in step.		Stack level, option and site attenuation are data and arithmetic evaluated on every call.
Site limits that depend on the chosen model and unit count: voltage availability, electrical service, roof weight, six chillers per controllerOnly with care
Each limit is its own rule against a different site attribute; a missed pairing lets a bad plant through.		Site attributes sit on the site record and are checked against the model and the unit count in one pass.
Lead time that changes with options (seismic and high-ambient kits) and discount authority that steps up with equipment valueOnly with care
Possible with option-dependent fields, but the lead time and the discount ceiling each need their own rule set kept in step with the price.		Lead time, the discount ceiling and the equipment total are derived on each call, and the line items add up to the total.
A stated total that does not match the computed priceOnly with care
A validation rule can compare a stated total with a calculated field, but only if the stated total is a field on the quote, which it rarely is.		The stated total travels with the quote and is compared with the computed one on every call.

  

  

    
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.

    
1Chiller model not offered at the voltage available at the site.
2More than 6 chillers on one plant controller.
3Minimum circuit ampacity above the electrical service at the site.
4Operating weight above the roof limit.
5Plant capacity at the design ambient and fluid, with any standby unit idle, below the load.
6Sound at the property line above the site limit.
7Winter minimum below 32F on plain water with no freeze protection kit.
8Seismic zone without seismic certification kits.
9Design ambient above 105F without the high-ambient kit.
10Lead time more than two weeks longer than the job allows.
11Discount above what anyone can approve (20 percent, or 28 percent on $1,000,000 or more of equipment).
12The agent's stated total does not match the computed total.
13Coastal site with uncoated coils (approval, engineering sign-off).
14Plant capacity above 1.5 times the load (approval, engineering confirmation).
15Discount above the 10 percent a sales rep can give (approval).
16Lead time one or two weeks longer than the job allows (approval, factory expedite).

  

  

    
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. Freeze protection: Chicago office tower plant on plain water

    
The agent's request. Please quote a chiller plant for Lakeshore Properties at the Chicago office tower: 280 tons of cooling from 3 AquaSnap 30RC100 chillers on plain water, no standby unit, needed in 16 weeks. The quote states a total of $379,500.00. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Freeze protection missing. Chicago office tower sees -10F in winter. An outdoor chiller on plain water needs a freeze protection kit, or use 30 percent glycol.
    • 
    

    
2. Freeze protection: The agent resubmits after fixing the errors

    
The agent's request. Please quote a chiller plant for Lakeshore Properties at the Chicago office tower: 280 tons of cooling from 3 AquaSnap 30RC100 chillers on plain water with freeze protection kits, no standby unit, needed in 16 weeks. The quote states a total of $393,000.00. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Quote summary. Total $393,000.00: chillers $354,000.00; options $13,500.00; services $10,500.00; controls $15,000.00.
    • 
    

    
3. High-ambient sites: Phoenix hospital plant at 116F design ambient

    
The agent's request. Please quote a chiller plant for Saguaro Health at the Phoenix hospital: 260 tons of cooling from 4 AquaSnap 30RC120 chillers on plain water with one unit on standby, needed in 18 weeks. The quote states a total of $581,000.00. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • High-ambient kit required. The 116F design ambient at Phoenix hospital is above 105F; the high-ambient kit is required.
    • 
    

    
4. High-ambient sites: The agent resubmits after fixing the errors

    
The agent's request. Please quote a chiller plant for Saguaro Health at the Phoenix hospital: 260 tons of cooling from 4 AquaSnap 30RC120 chillers on plain water with high-ambient kits and one unit on standby, needed in 18 weeks. The quote states a total of $609,000.00. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Quote summary. Total $609,000.00: chillers $552,000.00; options $28,000.00; services $14,000.00; controls $15,000.00.
    • 
    

    
5. Seismic and coastal sites: Los Angeles data center in a seismic zone on the coast

    
The agent's request. Please quote a chiller plant for Pacific Datavault at the Los Angeles data center: 400 tons of cooling from 4 AquaSnap 30RC150 chillers on plain water with one unit on standby, needed in 20 weeks. The quote states a total of $689,000.00. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Seismic kits required. Los Angeles data center is in a seismic zone; each chiller needs the seismic certification kit.
    • 
      
  • Coastal corrosion review. Los Angeles data center is a coastal site and the coils are uncoated. Corrosion voids coil warranty terms; needs engineering sign-off.
    • 
    

    
6. Seismic and coastal sites: The agent resubmits after fixing the errors

    
The agent's request. Please quote a chiller plant for Pacific Datavault at the Los Angeles data center: 400 tons of cooling from 4 AquaSnap 30RC150 chillers on plain water with seismic certification kits and one unit on standby, needed in 20 weeks. The quote states a total of $713,000.00. It is ready to send.

    
Verdict: Needs approval.

    
    
      
  • Coastal corrosion review. Los Angeles data center is a coastal site and the coils are uncoated. Corrosion voids coil warranty terms; needs engineering sign-off.
    • 
      
  • Quote summary. Total $713,000.00: chillers $660,000.00; options $24,000.00; services $14,000.00; controls $15,000.00.
    • 
    

    
7. Seismic and coastal sites: The agent resubmits after fixing the errors

    
The agent's request. Please quote a chiller plant for Pacific Datavault at the Los Angeles data center: 400 tons of cooling from 4 AquaSnap 30RC150 chillers on plain water with seismic certification kits and coil coating and one unit on standby, needed in 20 weeks. The quote states a total of $746,000.00. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Quote summary. Total $746,000.00: chillers $660,000.00; options $33,000.00; options $24,000.00; services $14,000.00; controls $15,000.00.
    • 
    

    
8. Stated total: Toronto school plant with a stated total

    
The agent's request. Please quote a chiller plant for the Maple Grove School Board at the Toronto school: 220 tons of cooling from 4 AquaSnap 30RC060 chillers on 30 percent glycol with the low-sound fan package, no standby unit, needed in 16 weeks. The quote states a total of $350,960.00. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Quoted total does not match. Stated total does not match: the agent stated $350,960.00; the validator computes $365,960.00.
    • 
    

  

  

    
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 drafts a quote for an air-cooled chiller plant: capacity at design conditions, sound, electrical load, and more. Before the quote goes out, the agent sends it to Predictika, which checks it against a declarative model, compares the total the agent states with the one it computes, and returns every finding together with the line-item quote.
    2. 
      
  2. I will start with the first request by the agent.
    3. 
      
  3. The agent says. Please quote a chiller plant for Lakeshore Properties at the Chicago office tower: 280 tons of cooling from 3 AquaSnap 30RC100 chillers on plain water. The quote states a total of $379,500.00. The rest of the request is on screen.
    4. 
      
  4. Before it goes out, the agent calls the validator.
    5. 
      
  5. Blocked, with one finding, all in one answer. Chicago office tower sees -10F in winter. An outdoor chiller on plain water needs a freeze protection kit, or use 30 percent glycol.
    6. 
      
  6. The validator returned one blocking finding 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. 
      
  7. What this shows: for site limits that depend on the chosen model and unit count: voltage availability, electrical service, roof weight, six chillers per controller, a rules-based CPQ is error-prone for this.
    8. 
      
  8. Now the agent resubmits.
    9. 
      
  9. The corrected request. It changes the quoted total to $393,000. It sets the freeze protection kits to yes.
    10. 
      
  10. It calls the validator again.
    11. 
      
  11. This time the validator clears it.
    12. 
      
  12. The validator also returns the line-item quote. It starts with AquaSnap 30RC100 and Freeze protection kit; the other two lines are on screen.
    13. 
      
  13. The earlier finding is cleared, and the validator's total, $393,000.00, matches the quote.
    14. 
      
  14. One problem caught in one call, fixed in one pass, and a clean $393,000 quote. That is the value of Predictika's validator: catching mistakes before they reach the customer. Now you can try the other scenarios yourself.
    15.