Predictika · Specialty polymer orders

An AI sales agent's PEEK resin order, checked before it goes out

An AI sales agent prepares an order for Syensqo KetaSpire PEEK: the grade, filler, form, color, number of 25 kg bags, the temperature the application needs, food-contact use and the total weight the order states. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of how those choices fit together and returns a verdict and every reason.

Scope: One KetaSpire PEEK order: grade, filler, form, color, bags, use temperature, food contact and the stated weight. See everything this validator covers

This validator models KetaSpire PEEK resin orders, not just the scenarios shown below.

  • Grades and fillers — KT-820, KT-880 and XT-920, each neat or with a GF30, CF30 or SL30 filler, with some fillers allowed on some grades only.
  • Form and color — pellets or fine powder, in natural or black, where fine powder is natural and neat only and carbon-filled grades are black.
  • Use temperature — a continuous-use limit for each grade, compared with what the application needs.
  • Quantity — 25 kg bags, a sample limit of 2 bags, supply-allocation approval above 80 bags and a block above 400 bags.
  • Food contact — not offered with carbon-filled or SL30 grades, and any other food-contact order goes to a person for a compliance check.
  • Stated weight — the weight the order states has to match the bags ordered.

Any KetaSpire PEEK order that touches these rules is in scope — not only the scenarios pictured here. Not covered: other Syensqo polymers and grades. Syensqo publishes no ordering rules for these choices, so every limit, threshold and rule beyond the public grade names is modeled. No prices are modeled.

The AI sales agentBuilds the PEEK resin order from the conversation with the customer 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 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
Fillers that fit some grades onlyOnly with care
The SL30 filler is modeled on KT-820 only, and fine powder is neat only, so each choice needs a rule that reads the grade and the filler together.		The grade, filler and form are fields of one request and are checked against each other on every call.
Colors that depend on the filler and the formOnly with care
Carbon-filled grades are black and fine powder is natural, so the color rule has to read both the filler and the form.		The color is checked against the filler and the form in the same pass on every call.
A temperature limit that depends on the gradeOnly with workarounds
XT-920 runs hotter than the other grades, so the limit has to be looked up from the grade and then compared with what the application needs.		The limit is computed from the grade and compared with the requirement in the same pass on every call.
A stated weight that must match the bagsOnly with workarounds
The weight on the order has to equal the bags times 25 kg, so a computed total has to be compared with a stated figure.		The total is computed in the same pass and compared with the stated weight on every call.
Sample, allocation and food-contact limitsOnly with workarounds
A sample limit, an approval band and a hard limit on quantity, plus a food-contact review, have to be kept in step with one another.		Blocks and approvals are evaluated together, and the verdict reports each one.
Every problem returned in one answerOnly with workarounds
Rules tend to fire as the order 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 order; approval findings need a person to sign off; everything else is computed.

    
1Fine powder is modeled for neat (unfilled) grades only; the order asks for a filled grade as powder.
2KT-880 is not offered as fine powder in this model; use KT-820 or XT-920.
3The SL30 wear-resistant filler is modeled on KT-820 only.
4Carbon and SL30 grades are black; natural color is not available.
5Fine powder is supplied natural only.
6The application needs 241 C continuous use; the selected grade is modeled to 240 C.
7Food-contact use is not modeled for carbon-filled or SL30 grades.
8Food-contact use needs a regulatory compliance check by a person.
9A sample order is limited to 2 bags; the order has 3.
10An order of 81 bags needs supply-allocation sign-off.
11An order of 401 bags is over the 400-bag (10,000 kg) single-order limit.
12The order states 400 kg; the bags ordered give 500 kg.

  

  

    
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. Grade, filler and form: A wear-resistant grade ordered as powder, over the temperature limit

    
The agent's request. Please quote KetaSpire KT-880 SL30 wear-resistant, as fine powder in black, 30 bags of 25 kg, as a production order. The application runs continuously at 250 C. The order states 700 kg in total. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Filled grade as powder. Fine powder is modeled for neat (unfilled) grades only; the order asks for a filled grade as powder.
    • 
      
  • KT-880 powder. KT-880 is not offered as fine powder in this model; use KT-820 or XT-920.
    • 
      
  • SL30 only on KT-820. The SL30 wear-resistant filler is modeled on KT-820 only.
    • 
      
  • Powder is natural only. Fine powder is supplied natural only.
    • 
      
  • Use temperature over the grade limit. The application needs 250 C continuous use; the selected grade is modeled to 240 C.
    • 
      
  • Stated weight does not match. The order states 700 kg; the bags ordered give 750 kg.
    • 
    

    
2. Grade, filler and form: The agent resubmits after fixing the errors

    
The agent's request. Please quote KetaSpire XT-920 neat (unfilled), as pellets in natural, 30 bags of 25 kg, as a production order. The application runs continuously at 250 C. The order states 750 kg in total. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Order summary. 30 bag(s) of 25 kg, 750 kg in total.
    • 
    

    
3. Food contact and volume: A carbon-filled natural grade for food contact, in a large order

    
The agent's request. Please quote KetaSpire XT-920 CF30 carbon-filled, as pellets in natural, 100 bags of 25 kg, as a production order. The application runs continuously at 245 C, and the resin will be in food contact. The order states 2500 kg in total. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Carbon filled grades are black. Carbon and SL30 grades are black; natural color is not available.
    • 
      
  • Food contact with carbon or wear filler. Food-contact use is not modeled for carbon-filled or SL30 grades.
    • 
      
  • Large order needs allocation. An order of 100 bags needs supply-allocation sign-off.
    • 
    

    
4. Food contact and volume: The agent resubmits after fixing the errors

    
The agent's request. Please quote KetaSpire XT-920 GF30 glass-filled, as pellets in natural, 100 bags of 25 kg, as a production order. The application runs continuously at 245 C. The order states 2500 kg in total. It is ready to send.

    
Verdict: Needs approval.

    
    
      
  • Large order needs allocation. An order of 100 bags needs supply-allocation sign-off.
    • 
      
  • Order summary. 100 bag(s) of 25 kg, 2,500 kg in total.
    • 
    

    
5. Sample order: A sample order that is too large and too hot for its grade

    
The agent's request. Please quote KetaSpire KT-820 GF30 glass-filled, as pellets in natural, 4 bags of 25 kg, as a sample order. The application runs continuously at 245 C. The order states 100 kg in total. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Use temperature over the grade limit. The application needs 245 C continuous use; the selected grade is modeled to 240 C.
    • 
      
  • Sample order too large. A sample order is limited to 2 bags; the order has 4.
    • 
    

    
6. Sample order: The agent resubmits after fixing the errors

    
The agent's request. Please quote KetaSpire XT-920 GF30 glass-filled, as pellets in natural, 2 bags of 25 kg, as a sample order. The application runs continuously at 245 C. The order states 50 kg in total. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Order summary. 2 bag(s) of 25 kg, 50 kg in total.
    • 
    

  

  

    
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 an order for Syensqo KetaSpire PEEK: the grade, filler, form, and more. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of how those choices fit together and returns a verdict and every reason.
    2. 
      
  2. I will start with the first request by the agent.
    3. 
      
  3. The agent says. Please quote KetaSpire KT-880 SL30 wear-resistant, as fine powder in black, 30 bags of 25 kg, as a production order. The application runs continuously at 250 C. The rest of the request is on screen.
    4. 
      
  4. Before it goes out, the agent calls the validator.
    5. 
      
  5. Blocked, with six findings, all in one answer. Fine powder is modeled for neat (unfilled) grades only; the order asks for a filled grade as powder. KT-880 is not offered as fine powder in this model; use KT-820 or XT-920. The SL30 wear-resistant filler is modeled on KT-820 only. The other three are on screen.
    6. 
      
  6. The validator returned six 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 order are checked against each other.
    7. 
      
  7. What this shows: for a temperature limit that depends on the grade, a rules-based CPQ can do this only with workarounds.
    8. 
      
  8. Now the agent resubmits.
    9. 
      
  9. The corrected request. It changes the grade to XT-920, the filler to neat, the form to pellets, the color to natural and the total weight stated on the order to 750 kg.
    10. 
      
  10. It calls the validator again.
    11. 
      
  11. This time the validator clears it.
    12. 
      
  12. All six earlier findings are cleared.
    13. 
      
  13. Six problems caught in one call, fixed in one pass, and a clean order. That is the value of Predictika's validator: catching mistakes before they reach the customer. Now you can try the other scenarios yourself.
    14.