Predictika · crusher sizing, wear-part fit, screen capacity and site power

An AI sales agent's cone crusher circuit quote, checked before it goes out

An AI sales agent prepares a crushing circuit quote for a fictional quarry: a cone crusher, a liner set, a screen, a rock type, a feed size, a closed side setting, a throughput, the site power and a number of circuits. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the sizing and wear-part rules and returns a verdict, every reason, and the price it computed.

Scope: Cone crusher circuit quotes in the style of Metso: three cone crushers, five liner sets and three screens, checked against sizing, wear-part fit, screen capacity, site power, quantity and discount rules. Prices and limits are illustrative estimates, not a vendor quote. See everything this validator covers

This validator models cone crusher circuits in the style of Metso, with three cone crushers, five liner sets and three screens, not just the scenarios shown below.

  • Crusher sizing — feed size, closed side setting and throughput against what each cone crusher accepts, with a review above 90 percent of its capacity, and the site power against the motor size.
  • Wear parts — whether the liner set covers the chosen closed side setting, whether abrasive rock (granite, iron ore) has an abrasion resistant liner set, and a review for abrasive rock at fine settings.
  • Screen capacity — whether the screen can handle the throughput the crusher is run at.
  • Circuits, discount and size — one to 6 circuits per quote, a volume discount from 2 circuits, an approval band above 5 percent total discount and a limit at 10 percent, and a deal desk review above USD 5,000,000.

Any quote for these crushers, liner sets, screens and rock types is in scope, not only the scenarios pictured here. Not checked: other Metso product lines such as grinding mills, mobile crushers or services, real list prices or lead times.

The AI sales agentBuilds the crushing circuit 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.

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

    
What the validator caughtA rules-based CPQPredictika
Sizing against several limitsOnly with workarounds
Feed, setting, throughput and power limits differ per crusher model, so each is a separate rule that must be edited with every new model.		All sizing limits are compared on every call from the crusher catalog.
Liner set against the settingOnly with care
An option filter on a customer-entered number that has to agree with the crusher range.		The setting is checked against both the crusher range and the liner range.
Abrasive rock needs abrasion resistant linersOnly with care
A dependency between an application choice and a part property that rules rarely hold.		The rock type is compared with the liner property directly.
Screen capacity against throughputOnly with workarounds
A value from one product compared with a value from another, usually left to a manual check.		Screen capacity and throughput are compared on every call.
Every problem returned in one answerOnly with workarounds
Rules tend to fire as the quote is saved, 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 feed size, the closed side setting and the throughput must be within what the cone crusher accepts.
2The site power must cover the motor of the cone crusher.
3The closed side setting must be within the range of the liner set.
4Abrasive rock such as granite or iron ore needs an abrasion resistant liner set.
5The screen must be able to handle the throughput of the circuit.
6A throughput above 90 percent of the crusher capacity needs application engineering to confirm.
7Abrasive rock at a closed side setting below 12 mm needs application engineering to confirm the wear life.
8A quote holds one to 6 circuits.
9A total discount above 5 percent needs the sales director, and above 10 percent is blocked.
10An order above USD 5,000,000 needs a deal desk review.
11The net total the agent states must equal the total the validator computes.

  

  

    
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. Liner set against the setting: Coarse liners at a 20 mm closed side setting

    
The agent's request. Please check a quote for one HP300 cone crusher with the Coarse manganese liner set and a three deck screen, on limestone with a 200 mm feed, a 20 mm closed side setting and 350 tons per hour, with 250 kW of site power, for our fictional quarry customer Ridgeline Aggregates. The quoted total is USD 1,345,000. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Liner range. A closed side setting of 20 mm is outside the 25 to 52 mm of the Coarse manganese liner set.
    • 
      
  • Estimated price. The estimated list price is USD 1,345,000 for 1 circuit, with 0 percent total discount. Prices are illustrative estimates.
    • 
    

    
2. Liner set against the setting: The agent resubmits after fixing the error

    
The agent's request. Please check the quote again. For Ridgeline Aggregates I changed the liners to the Medium manganese liner set: one HP300 cone crusher with a three deck screen, on limestone with a 200 mm feed, a 20 mm closed side setting and 350 tons per hour, with 250 kW of site power. The quoted total is USD 1,342,000. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Estimated price. The estimated list price is USD 1,342,000 for 1 circuit, with 0 percent total discount. Prices are illustrative estimates.
    • 
    

    
3. Discount over the limit: Two circuits with 9 percent extra discount

    
The agent's request. Please check a quote for two HP300 circuits with the Medium manganese liner set and a three deck screen, on limestone with a 200 mm feed, a 20 mm closed side setting and 350 tons per hour, with 250 kW of site power, for Ridgeline Aggregates. I added 9 percent extra discount to win the order. The quoted total is USD 2,388,760. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Discount limit. The total discount of 11 percent exceeds the 10 percent limit.
    • 
      
  • Estimated price. The estimated list price is USD 2,388,760 for 2 circuits, with 11 percent total discount. Prices are illustrative estimates.
    • 
    

    
4. Discount over the limit: The agent resubmits after fixing the error

    
The agent's request. Please check the quote again. For Ridgeline Aggregates I cut the extra discount to 2 percent: two HP300 circuits with the Medium manganese liner set and a three deck screen, on limestone with a 200 mm feed, a 20 mm closed side setting and 350 tons per hour, with 250 kW of site power. The quoted total is USD 2,576,640. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Estimated price. The estimated list price is USD 2,576,640 for 2 circuits, with 4 percent total discount. Prices are illustrative estimates.
    • 
    

    
5. Quoted total that does not match: A quoted total below the price the validator computes

    
The agent's request. Please check a quote for one HP200 cone crusher with the Medium manganese liner set and a two deck screen, on limestone with a 150 mm feed, a 16 mm closed side setting and 200 tons per hour, with 140 kW of site power, for Ridgeline Aggregates. The quoted total is USD 838,860. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Quoted total. The quoted net total of USD 838,860 does not match the USD 902,000 the validator computed.
    • 
      
  • Estimated price. The estimated list price is USD 902,000 for 1 circuit, with 0 percent total discount. Prices are illustrative estimates.
    • 
    

    
6. Quoted total that does not match: The agent resubmits after fixing the error

    
The agent's request. Please check the quote again. For Ridgeline Aggregates I corrected the total: one HP200 cone crusher with the Medium manganese liner set and a two deck screen, on limestone with a 150 mm feed, a 16 mm closed side setting and 200 tons per hour, with 140 kW of site power. The quoted total is USD 902,000. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Estimated price. The estimated list price is USD 902,000 for 1 circuit, with 0 percent total discount. Prices are illustrative estimates.
    • 
    

  

  

    
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 crushing circuit quote for a fictional quarry: a cone crusher, a liner set, a screen, and more. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the sizing and wear-part rules and returns a verdict, every reason, and the price it computed.
    2. 
      
  2. I will start with the first request by the agent.
    3. 
      
  3. The agent says. Please check a quote for two HP300 circuits with the Medium manganese liner set and a three deck screen, on limestone with a 200 mm feed. I added 9 percent extra discount to win the order. 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. The total discount of 11 percent exceeds the 10 percent limit.
    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 sizing against several limits, a rules-based CPQ can do this only with workarounds.
    8. 
      
  8. Now the agent resubmits.
    9. 
      
  9. The corrected request. It changes the extra discount requested to 2 percent.
    10. 
      
  10. It calls the validator again.
    11. 
      
  11. This time the validator clears it.
    12. 
      
  12. The earlier finding is cleared.
    13. 
      
  13. One problem 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.
    14.