Predictika · 3D radar orders

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

An AI sales agent prepares an order for Indra LANZA 3D radars with a model, a deployment, a range, options, spares, training and a support period. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the product family and returns a verdict, every reason and the total it computed.

Scope: Indra LANZA 3D radars (LRR, LTR-25, LTR-20, 3D PSR, LANZA N) against range class, deployment, IFF Mode 5, TBM detection, civil air traffic control use, console, spares, support, discount and total. See everything this validator covers

This validator models Indra LANZA radar orders, not just the scenarios shown below.

  • Model, range and deployment — the range class each model reaches and whether it is offered fixed, tactical or shipborne.
  • Options — IFF Mode 5, tactical ballistic missile detection, civil air traffic control use and the radar control console, each available on some models only.
  • Spares, support, discount and total — spare kits per radar, the support tier and period, approval steps for larger discounts and for programme-size orders, a discount ceiling, and the order total against the total computed from the modeled prices.

Any Indra LANZA radar order that touches these rules is in scope — not only the scenarios pictured here. Not covered: other Indra radar families and export licensing. The family members, range classes and main capabilities come from Indra's public brochures; every other rule and all prices are modeled, not published by Indra.

The AI sales agentBuilds the radar 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
A range that differs for each modelOnly with care
Each model reaches its own range class, so each needs its own limit rule.		The model and the range needed are fields of one request and compared on every call.
Deployments and options that only some models offerOnly with care
Fixed, tactical and shipborne use, Mode 5, TBM detection and civil ATC use each depend on the model, so the pairings multiply into many rules.		The model, deployment and options are fields of one request and checked together.
Support tiers with their own period limitOnly with care
Basic support is limited to three years while the other tiers run longer, so the tier and the period need a joint rule.		Both are fields of the request and compared on every call.
A stated total that must match the computed totalOnly with workarounds
The total is built from the model, options, spares, training, support years and discount, so a separate rule is needed to recompute it and compare it with the stated figure.		The stated total is a field of the request and is compared with the computed total on every call.
Every problem returned in one answerOnly with workarounds
Rules tend to fire as a 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 customer needs 201 NM of instrumented range; this model reaches 200 NM.
2The customer needs 251 NM of instrumented range; this model reaches 250 NM.
3This radar model is not offered in the selected deployment.
4The 3D PSR air traffic control radar is not offered with IFF Mode 5.
5IFF Mode 5 is a controlled option; export and end-user checks must be cleared by a person.
6Tactical ballistic missile detection is offered only on the LRR and the LTR-25.
7This radar model is not offered for civil air traffic control use.
8Civil air traffic control use of a defence-class radar needs aviation authority and programme confirmation.
9The Local/Remote Radar Control Console is a ground option and cannot be quoted for a shipborne radar.
10Basic support is sold for up to 3 years; the order asks for 4.
11A support period of 11 years is outside 1 to 10 years.
125 spare kits is more than 2 per radar for the order.
13A discount of 12 percent is above the 8 percent a sales rep can give.
14A discount of 16 percent is above the 15 percent limit.
15An order of 4 radars is a programme deal that sales management reviews.
16The order states a total of 34,601,000 euros; the modeled prices give 34,600,000 euros.
17The order has 0 radars; at least 1 is needed.

  

  

    
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, deployment and support: An LTR-25 order for civil air traffic control with a long range, spares and Basic support

    
The agent's request. Please quote 2 LANZA LTR-25 radars, fixed deployment, 260 NM of range, civil air traffic control use, 5 spare kits, Basic support for 4 years and a 16 percent discount, for a total of 46,620,000 euros. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Range above the model class. The customer needs 260 NM of instrumented range; this model reaches 250 NM.
    • 
      
  • Deployment not offered for the model. This radar model is not offered in the selected deployment.
    • 
      
  • Civil air traffic control use not offered. This radar model is not offered for civil air traffic control use.
    • 
      
  • Basic support limited to three years. Basic support is sold for up to 3 years; the order asks for 4.
    • 
      
  • Too many spare kits. 5 spare kits is more than 2 per radar for the order.
    • 
      
  • Discount above 15 percent. A discount of 16 percent is above the 15 percent limit.
    • 
    

    
2. Model, deployment and support: The agent resubmits after fixing the errors

    
The agent's request. Please quote 2 LANZA 3D PSR radars, fixed deployment, 150 NM of range, civil air traffic control use, 4 spare kits, Basic support for 3 years and an 8 percent discount, for a total of 22,724,000 euros. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Order summary. 2 radar(s) up to 150 NM; list subtotal 24,700 thousand euros; order total after discount 22,724,000 euros (modeled prices).
    • 
    

    
3. Naval radar options: A shipborne LANZA N order with a console, TBM detection and a long range

    
The agent's request. Please quote 1 LANZA N radar, shipborne deployment, 200 NM of range, TBM detection, the remote radar control console, Standard support for 3 years and a 0 percent discount, for a total of 23,045,000 euros. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Range above the model class. The customer needs 200 NM of instrumented range; this model reaches 180 NM.
    • 
      
  • TBM detection not offered. Tactical ballistic missile detection is offered only on the LRR and the LTR-25.
    • 
      
  • Remote console on a naval radar. The Local/Remote Radar Control Console is a ground option and cannot be quoted for a shipborne radar.
    • 
      
  • Stated total does not match. The order states a total of 23,045,000 euros; the modeled prices give 23,050,000 euros.
    • 
    

    
4. Naval radar options: The agent resubmits after fixing the errors

    
The agent's request. Please quote 1 LANZA N radar, shipborne deployment, 180 NM of range, Standard support for 3 years and a 0 percent discount, for a total of 20,800,000 euros. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Order summary. 1 radar(s) up to 180 NM; list subtotal 20,800 thousand euros; order total after discount 20,800,000 euros (modeled prices).
    • 
    

    
5. Programme-size order: Four radars with Mode 5, civil air traffic control use and a larger discount

    
The agent's request. Please quote 4 LANZA LTR-20 radars, tactical deployment, 120 NM of range, IFF Mode 5, civil air traffic control use, Standard support for 3 years and a 12 percent discount, for a total of 65,824,000 euros. It is ready to send.

    
Verdict: Needs approval.

    
    
      
  • Civil use of a defence radar. Civil air traffic control use of a defence-class radar needs aviation authority and programme confirmation.
    • 
      
  • Mode 5 export check. IFF Mode 5 is a controlled option; export and end-user checks must be cleared by a person.
    • 
      
  • Discount above 8 percent. A discount of 12 percent is above the 8 percent a sales rep can give.
    • 
      
  • Four or more radars. An order of 4 radars is a programme deal that sales management reviews.
    • 
      
  • Order summary. 4 radar(s) up to 200 NM; list subtotal 74,800 thousand euros; order total after discount 65,824,000 euros (modeled prices).
    • 
    

    
6. Programme-size order: The agent resubmits after fixing the errors

    
The agent's request. Please quote 3 LANZA LTR-20 radars, tactical deployment, 120 NM of range, Standard support for 3 years and a 6 percent discount, for a total of 48,786,000 euros. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Order summary. 3 radar(s) up to 200 NM; list subtotal 51,900 thousand euros; order total after discount 48,786,000 euros (modeled prices).
    • 
    

  

  

    
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 Indra LANZA 3D radars with a model, a deployment, a range, options, spares, training and a support period. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the product family and returns a verdict, every reason and the total it computed.
    2. 
      
  2. I will start with the first request by the agent.
    3. 
      
  3. The agent says. Please quote 2 LANZA LTR-25 radars, fixed deployment, 260 NM of range, civil air traffic control use, 5 spare kits. It is ready to send. 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. The customer needs 260 NM of instrumented range; this model reaches 250 NM. This radar model is not offered in the selected deployment. This radar model is not offered for civil air traffic control use. 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 quote are checked against each other.
    7. 
      
  7. What this shows: for a range that differs for each model, a rules-based CPQ is error-prone for this.
    8. 
      
  8. Now the agent resubmits.
    9. 
      
  9. The corrected request. It changes the radar model to LANZA 3D PSR, the range needed to 150 NM, the spare kits to 4, the support period to 3 years, the discount to 8 percent and the stated total to €22,724,000.
    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 22,724,000 euros 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.