Predictika · Agricultural planters

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

An AI sales agent prepares an order for a Tatu Marchesan USAP or USAP Suprema grain planter for a fictional soybean and corn grower: the row count and row spacing, the line, the fertilizer and seed version, the seed the farm needs to carry, the transport route and the tractor. Before the order goes out, the agent sends it to Predictika, which checks it against a declarative model of what the public Marchesan USAP brochure says and returns a verdict, every reason and the figures it computed. The customer is fictional.

Scope: Marchesan USAP and USAP Suprema planters (18 to 50 rows) only: row count and spacing, line, fertilizer version, working width, seed capacity, transport size and tractor power. See everything this validator covers

This validator models orders for the Tatu Marchesan USAP and USAP Suprema grain planters, based on the public Marchesan USAP brochure (06/2019, Rev. 06), not just the scenarios shown below.

  • Row count and spacing — the row counts in the brochure table (18 to 50) and the 450, 500 and 550 mm row spacings; other row counts go to a person, because the brochure says to consult the factory.
  • Fertilizer and seed version — listed under the USAP with up to 30 rows; on the Suprema it goes to a person.
  • Working width and seed capacity — the working width and seed hopper capacity the brochure gives for the row count, compared with the width the order states and the seed the farm needs to carry.
  • Transport — the folded width and length the brochure gives, against the limits of the transport route.
  • Tractor power — the required power range the brochure gives for each row count and line: below the minimum is blocked; inside the range goes to a person to confirm for the soil and field.

Any order of a USAP or USAP Suprema planter that touches these rules is in scope — not only the scenarios pictured here. Not covered: other Marchesan machines, metering and crop choices, prices (Marchesan does not publish them), delivery, financing and service.

The AI sales agentBuilds the planter 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.

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
Figures that depend on the row countOnly with care
Working width, seed capacity, transport length and tractor power all follow the row count in the brochure table. A rules-based CPQ can store the table, but it has to be kept in step and compared with each order.		The row count and line are read together and every figure is computed on every call.
A limit that applies to only some versionsOnly with care
The fertilizer and seed version is listed up to 30 rows under the USAP. Each is a product rule that has to be written and maintained.		The version limit is checked together with the row count and line on every call.
The figure the order statesOnly with care
The working width stated in the order is a free-entry field; a rule has to look up the right table row and compare.		The stated width is a field in the request, compared with the brochure figure for the row count on every call.
Every problem returned in one answerOnly with workarounds
Rules tend to fire as the quote is saved or a step is completed, so a person 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 brochure lists row spacings of 450, 500 and 550 mm; 600 mm is not listed.
2The brochure lists row spacings of 450, 500 and 550 mm; 475 mm is not listed.
3The brochure lists the USAP fertilizer and seed version with up to 30 rows; the order has 32 rows.
4The brochure lists the USAP fertilizer and seed version with up to 30 rows; the order has 36 rows.
5The brochure lists the fertilizer and seed version under the USAP; a person should confirm it for the USAP Suprema.
6The brochure table has no entry for 19 rows and says to consult the factory for other configurations; a person should confirm it.
7The brochure table has no entry for 51 rows and says to consult the factory for other configurations; a person should confirm it.
8The brochure table has no entry for 10 rows and says to consult the factory for other configurations; a person should confirm it.
9The brochure table has no entry for 31 rows and says to consult the factory for other configurations; a person should confirm it.
10The tractor has 279 cv; the brochure gives a required power of at least 280 cv for this planter.
11The tractor has 280 cv, inside the 280 to 300 cv range the brochure gives; a person should confirm it for the soil and field.
12The tractor has 299 cv, inside the 280 to 300 cv range the brochure gives; a person should confirm it for the soil and field.
13The tractor has 300 cv; the brochure gives a required power of at least 320 cv for this planter.
14The tractor has 300 cv, inside the 290 to 320 cv range the brochure gives; a person should confirm it for the soil and field.
15The tractor has 200 cv; the brochure gives a required power of at least 210 cv for this planter.
16The farm needs a seed capacity of 4681 liters; this planter holds 4680 liters.
17The farm needs a seed capacity of 5000 liters; this planter holds 4160 liters.
18The transport route allows 6499 mm; the planter is 6500 mm wide folded for transport.
19The transport route allows 12299 mm; the planter is 12300 mm long folded for transport.
20The transport route allows 18000 mm; the planter is 18600 mm long folded for transport.

  

  

    
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. Tractor power: USAP Suprema, 36 rows, with a 300 cv tractor

    
The agent's request. Please quote a Marchesan USAP Suprema planter with 36 rows at 500 mm spacing, the standard version. The farm needs to carry 3,000 liters of seed, the transport route allows 7,000 mm wide and 20,000 mm long when folded, and the tractor is 300 cv. The quote states a working width of 17,800 mm. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Tractor too small. The tractor has 300 cv; the brochure gives a required power of at least 320 cv for this planter.
    • 
    

    
2. Tractor power: The agent resubmits after fixing the errors

    
The agent's request. Please quote a Marchesan USAP planter with 36 rows at 500 mm spacing, the standard version. The farm needs to carry 3,000 liters of seed, the transport route allows 7,000 mm wide and 20,000 mm long when folded, and the tractor is 300 cv. The quote states a working width of 17,800 mm. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Order summary. This planter works 17800 mm wide, holds 6240 liters of seed, weighs about 18234 kg and needs a tractor of 280 to 300 cv.
    • 
    

    
3. Fertilizer and seed version: Fertilizer and seed version with 36 rows

    
The agent's request. Please quote a Marchesan USAP planter with 36 rows at 500 mm spacing, the fertilizer and seed version. The farm needs to carry 3,000 liters of seed, the transport route allows 7,000 mm wide and 20,000 mm long when folded, and the tractor is 300 cv. The quote states a working width of 17,800 mm. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Fertilizer version limited to 30 rows. The brochure lists the USAP fertilizer and seed version with up to 30 rows; the order has 36 rows.
    • 
    

    
4. Fertilizer and seed version: The agent resubmits after fixing the errors

    
The agent's request. Please quote a Marchesan USAP planter with 30 rows at 500 mm spacing, the fertilizer and seed version. The farm needs to carry 3,000 liters of seed, the transport route allows 7,000 mm wide and 20,000 mm long when folded, and the tractor is 300 cv. The quote states a working width of 14,800 mm. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Order summary. This planter works 14800 mm wide, holds 4680 liters of seed, weighs about 15867 kg and needs a tractor of 230 to 250 cv.
    • 
    

    
5. Seed capacity: 28 rows for a farm needing 5,000 liters of seed

    
The agent's request. Please quote a Marchesan USAP planter with 28 rows at 500 mm spacing, the standard version. The farm needs to carry 5,000 liters of seed, the transport route allows 7,000 mm wide and 20,000 mm long when folded, and the tractor is 300 cv. The quote states a working width of 13,800 mm. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Seed hopper too small. The farm needs a seed capacity of 5000 liters; this planter holds 4160 liters.
    • 
    

    
6. Seed capacity: The agent resubmits after fixing the errors

    
The agent's request. Please quote a Marchesan USAP planter with 33 rows at 500 mm spacing, the standard version. The farm needs to carry 5,000 liters of seed, the transport route allows 7,000 mm wide and 20,000 mm long when folded, and the tractor is 300 cv. The quote states a working width of 16,300 mm. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Order summary. This planter works 16300 mm wide, holds 5720 liters of seed, weighs about 17360 kg and needs a tractor of 250 to 270 cv.
    • 
    

    
7. Row spacing: 28 rows at 600 mm spacing

    
The agent's request. Please quote a Marchesan USAP planter with 28 rows at 600 mm spacing, the standard version. The farm needs to carry 3,000 liters of seed, the transport route allows 7,000 mm wide and 20,000 mm long when folded, and the tractor is 300 cv. The quote states a working width of 13,800 mm. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Row spacing not offered. The brochure lists row spacings of 450, 500 and 550 mm; 600 mm is not listed.
    • 
    

    
8. Row spacing: The agent resubmits after fixing the errors

    
The agent's request. Please quote a Marchesan USAP planter with 28 rows at 550 mm spacing, the standard version. The farm needs to carry 3,000 liters of seed, the transport route allows 7,000 mm wide and 20,000 mm long when folded, and the tractor is 300 cv. The quote states a working width of 13,800 mm. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Order summary. This planter works 13800 mm wide, holds 4160 liters of seed, weighs about 14727 kg and needs a tractor of 210 to 230 cv.
    • 
    

  

  

    
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 a Tatu Marchesan USAP or USAP Suprema grain planter for a fictional soybean and corn grower: the row count and row spacing, the line, the fertilizer and seed version, and more. Before the order goes out, the agent sends it to Predictika, which checks it against a declarative model of what the public Marchesan USAP brochure says and returns a verdict, every reason and the figures it computed. The customer is fictional.
    2. 
      
  2. I will start with the first request by the agent.
    3. 
      
  3. The agent says. Please quote a Marchesan USAP Suprema planter with 36 rows at 500 mm spacing, the standard version. The farm needs to carry 3,000 liters of seed, the transport route allows 7,000 mm wide and 20,000 mm long when folded, and the tractor is 300 cv. The rest of the request is on screen. The quote states a working width of 17,800 mm.
    4. 
      
  4. Before it goes out, the agent calls the validator.
    5. 
      
  5. Blocked, with one finding, all in one answer. The tractor has 300 cv; the brochure gives a required power of at least 320 cv for this planter.
    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 figures that depend on the row count, a rules-based CPQ is error-prone for this.
    8. 
      
  8. Now the agent resubmits.
    9. 
      
  9. The corrected request. It changes the line to USAP.
    10. 
      
  10. It calls the validator again.
    11. 
      
  11. This time the validator clears it.
    12. 
      
  12. The earlier finding is cleared, and the validator's figure matches the quote.
    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.