Predictika · Distribution transformer orders

An AI sales agent's distribution transformer order, checked before it goes out

An AI sales agent prepares an order for an oil-immersed or dry-type distribution transformer: rating, highest voltage, winding options, the loss levels it promises and, for dry-type, the test voltages and thermal alarm. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the published KONČAR documents and returns a verdict, every reason and the figures it looked up.

Scope: Oil-immersed (50 to 8000 kVA, up to 36 kV) and dry-type distribution transformers: range, voltage, EU Ecodesign loss levels, efficiency index, winding options, and dry-type test voltages and thermal alarm. See everything this validator covers

This validator models distribution transformer orders against KONČAR's public Ecodesign brochure, distribution transformer brochure and dry-type brochure, not just the scenarios shown below.

  • Range and voltage — the rated power must be inside the published range for the type, and dry-type units are published up to 24 kV only; where KONČAR's documents disagree on the dry-type range, the order is sent for confirmation.
  • Ecodesign loss levels — the no-load and load loss classes and the minimum peak efficiency index for large units are checked for the tier on the order (Tier 1 or Tier 2), and the loss-limit uplifts the quote states must equal the published correction table for the winding option chosen.
  • Options and dry-type data — dual-voltage reduced power, winding options that must match the highest voltage, dry-type test voltages and thermal alarm temperature by insulation class, and special conditions (higher ambient or altitude, engineered-to-order applications) that need approval.

Any order for these transformers that touches these rules is in scope — not only the scenarios pictured here. Not covered: prices (KONČAR publishes no price list, so this validator produces no price build-up or line-item quote), the numeric loss limits in watts, which are in the EU regulation and not in the documents read, and power transformers above 8000 kVA.

The AI sales agentBuilds the transformer order 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 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
Loss class and tier together decide what is allowedOnly with workarounds
The no-load margin, the load class and the minimum efficiency all depend on type, rating band and tier at once, so each needs its own lookup keyed on the combination.		Type, rating and tier are in one request and the requirement is looked up from the combination on every call.
A stated uplift that must equal the correction tableOnly with care
The uplift is a free-entry figure; a rule can compare it with a looked-up value, but someone has to write and maintain that comparison for every winding option.		The stated figures travel with the order and are compared with the looked-up figures on every call.
Test voltages and alarm temperature that follow the voltage and insulation classOnly with care
The expected value depends on the highest voltage or the insulation class, so each check needs a lookup keyed on that field.		The expected test voltages and alarm temperature are looked up from the voltage and insulation class and compared with the quote.
Documents that disagree (dry-type range)Only with workarounds
A rules table normally holds one limit; two published limits need a middle band that is neither allowed nor refused.		The band between the two published limits is flagged for confirmation instead of being allowed or refused.
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 quote; approval findings need a person to sign off; everything else is computed.

    
1The quoted peak efficiency index is 99531 but the minimum for this rating and tier is 99532 (the index is given in thousandths of a percent).
2The quoted peak efficiency index is 99509 but the minimum for this rating and tier is 99510 (the index is given in thousandths of a percent).
3The quoted peak efficiency index is 99483 but the minimum for this rating and tier is 99548 (the index is given in thousandths of a percent).
4The rating lies between ratings in the peak efficiency table; the minimum is obtained by interpolation and is not computed here.
5A rated power of 8500 kVA is outside the published range for this transformer type (oil-immersed 50 to 8000 kVA; dry-type up to 5000 kVA).
6A rated power of 40 kVA is outside the published range for this transformer type (oil-immersed 50 to 8000 kVA; dry-type up to 5000 kVA).
7A rated power of 6300 kVA is outside the published range for this transformer type (oil-immersed 50 to 8000 kVA; dry-type up to 5000 kVA).
8KONČAR documents give the dry-type range as up to 3150 kVA on its website and up to 5000 kVA in its brochures, so a rating in between needs confirmation.
9Dry-type distribution transformers are published for a highest voltage for equipment up to 24 kV; 36 kV is for oil-immersed transformers.
10The quoted no-load losses do not meet the Ecodesign requirement for this type, rating and tier (class A0, and 10 percent below the A0 level from Tier 2).
11The quoted load losses are above the Ecodesign limit for this type, rating and tier, which is class Bk.
12The quoted load losses are above the Ecodesign limit for this type, rating and tier, which is class Ak.
13The quote states loss-limit uplifts of 20 percent (no-load) and 10 percent (load), but the Ecodesign correction table gives 20 and 15 percent.
14The quote states loss-limit uplifts of 20 percent (no-load) and 5 percent (load), but the Ecodesign correction table gives 10 and 5 percent.
15With dual voltage on one winding at reduced power, the maximum power on the lower voltage is 85 percent of the rating at the higher voltage; the quote asks for 100 percent.
16The selected winding feature does not match the highest voltage for equipment on the quote (the 36 kV rows apply to Um = 36 kV, the other row to Um of 24 kV or less).
17The quote states test voltages of 50 kV (power frequency) and 95 kV (lightning impulse), but the dry-type brochure gives 28 and 75 kV for this highest voltage for equipment.
18The quote sets the thermal-protection alarm at 135 C, but the dry-type brochure gives 160 C for this insulation temperature class.
19The dry-type brochure bases ratings on a 40 C maximum ambient and an altitude up to 1000 m; higher values must be specified in the enquiry and confirmed.
20Rectifier, furnace, offshore, earthing and railway transformers are engineered to order and are outside the EU Ecodesign Regulation; KONČAR publishes no ratings or options for them, so none are checked.
21The quote states test voltages of 50 kV (power frequency) and 125 kV (lightning impulse), but the dry-type brochure gives 50 and 95 kV for this highest voltage for equipment.
22The quote sets the thermal-protection alarm at 150 C, but the dry-type brochure gives 135 C for this insulation temperature class.

  

  

    
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. No-load loss level: A Tier 2 oil-immersed transformer quoted without the no-load margin

    
The agent's request. Please quote a 1000 kVA oil-immersed distribution transformer, highest voltage for equipment 24 kV, standard windings, Tier 2. The quote promises no-load losses at class A0 with a 0 percent margin and load losses at class Ak, and it states loss-limit uplifts of 0 percent (no-load) and 0 percent (load). It is ready to send.

    
Verdict: Blocked.

    
    
      
  • No-load loss class too weak. The quoted no-load losses do not meet the Ecodesign requirement for this type, rating and tier (class A0, and 10 percent below the A0 level from Tier 2).
    • 
    

    
2. No-load loss level: The agent resubmits after fixing the errors

    
The agent's request. Please quote a 1000 kVA oil-immersed distribution transformer, highest voltage for equipment 24 kV, standard windings, Tier 2. The quote promises no-load losses at class A0 with a 10 percent margin and load losses at class Ak, and it states loss-limit uplifts of 0 percent (no-load) and 0 percent (load). It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Looked up. Loss-limit uplifts for this order: no-load 0 percent, load 0 percent; minimum peak efficiency index 0 (0 = not applicable at this rating).
    • 
    

    
3. Stated loss-limit uplift: A 36 kV transformer whose stated load uplift is not the table figure

    
The agent's request. Please quote a 1000 kVA oil-immersed distribution transformer, highest voltage for equipment 36 kV, 36 kV primary with secondary over 1.1 kV, Tier 2. The quote promises no-load losses at class A0 with a 10 percent margin and load losses at class Ak, and it states loss-limit uplifts of 20 percent (no-load) and 10 percent (load). It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Stated loss uplifts differ from the table. The quote states loss-limit uplifts of 20 percent (no-load) and 10 percent (load), but the Ecodesign correction table gives 20 and 15 percent.
    • 
    

    
4. Stated loss-limit uplift: The agent resubmits after fixing the errors

    
The agent's request. Please quote a 1000 kVA oil-immersed distribution transformer, highest voltage for equipment 36 kV, 36 kV primary with secondary over 1.1 kV, Tier 2. The quote promises no-load losses at class A0 with a 10 percent margin and load losses at class Ak, and it states loss-limit uplifts of 20 percent (no-load) and 15 percent (load). It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Looked up. Loss-limit uplifts for this order: no-load 20 percent, load 15 percent; minimum peak efficiency index 0 (0 = not applicable at this rating).
    • 
    

    
5. Dry-type test data: A dry-type unit with the wrong thermal alarm and test voltages

    
The agent's request. Please quote a 800 kVA dry-type distribution transformer, highest voltage for equipment 12 kV, standard windings, Tier 2. The quote promises no-load losses at class A0 with a 10 percent margin and load losses at class Ak, and it states loss-limit uplifts of 0 percent (no-load) and 0 percent (load). The dry-type unit is insulation class H with the thermal alarm set at 135 C, and the quote gives test voltages of 50 kV power frequency and 95 kV lightning impulse. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Stated test voltages differ from the table. The quote states test voltages of 50 kV (power frequency) and 95 kV (lightning impulse), but the dry-type brochure gives 28 and 75 kV for this highest voltage for equipment.
    • 
      
  • Thermal alarm setting differs from the table. The quote sets the thermal-protection alarm at 135 C, but the dry-type brochure gives 160 C for this insulation temperature class.
    • 
    

    
6. Dry-type test data: The agent resubmits after fixing the errors

    
The agent's request. Please quote a 800 kVA dry-type distribution transformer, highest voltage for equipment 12 kV, standard windings, Tier 2. The quote promises no-load losses at class A0 with a 10 percent margin and load losses at class Ak, and it states loss-limit uplifts of 0 percent (no-load) and 0 percent (load). The dry-type unit is insulation class H with the thermal alarm set at 160 C, and the quote gives test voltages of 28 kV power frequency and 75 kV lightning impulse. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Looked up. Loss-limit uplifts for this order: no-load 0 percent, load 0 percent; minimum peak efficiency index 0 (0 = not applicable at this rating).
    • 
    

  

  

    
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 an oil-immersed or dry-type distribution transformer: rating, highest voltage, winding options, and more. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the published KONČAR documents and returns a verdict, every reason and the figures it looked up.
    2. 
      
  2. I will start with the first request by the agent.
    3. 
      
  3. The agent says. Please quote a 800 kVA dry-type distribution transformer, highest voltage for equipment 12 kV, standard windings, Tier 2. The quote promises no-load losses at class A0 with a 10 percent margin and load losses at class Ak, and it states loss-limit uplifts of 0 percent (no-load) and 0 percent (load). The rest of the request is on screen.
    4. 
      
  4. Before it goes out, the agent calls the validator.
    5. 
      
  5. Blocked, with two findings, all in one answer. The quote states test voltages of 50 kV (power frequency) and 95 kV (lightning impulse), but the dry-type brochure gives 28 and 75 kV for this highest voltage for equipment. The quote sets the thermal-protection alarm at 135 C, but the dry-type brochure gives 160 C for this insulation temperature class.
    6. 
      
  6. The validator returned two 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. 
      
  7. What this shows: for test voltages and alarm temperature that follow the voltage and insulation class, a rules-based CPQ is error-prone for this.
    8. 
      
  8. Now the agent resubmits.
    9. 
      
  9. The corrected request. It changes the thermal alarm to 160 C, the test voltage, power frequency to 28 kV and the test voltage, lightning impulse to 75 kV.
    10. 
      
  10. It calls the validator again.
    11. 
      
  11. This time the validator clears it.
    12. 
      
  12. Both earlier findings are cleared.
    13. 
      
  13. Two 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.
    14.