Predictika · Beverage can orders

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

An AI sales agent prepares an order for beverage cans: a can format and fill volume, an end size and end type, a tab, a decoration technology, the number of designs and a number of pallets, with a stated total of cans. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the Orora Beverage Cans range and returns a verdict, every reason and the can total it computed.

Scope: The Orora Beverage Cans Classic, Slim and Sleek formats with end sizes and types, tabs and decoration technologies, designs per run, minimum runs and whole-pallet quantities. See everything this validator covers

This validator models Orora beverage can orders, not just the scenarios shown below.

  • Format, fill and end — twelve can formats from 150 to 500 mL, the fill volume each holds, the 50 or 52 mm end each takes, and the easy-open, XO resealable and full aperture end types.
  • Tab and decoration — special tabs not on the resealable end, one design per run unless Flex FX (up to 16 designs), and minimum runs by decoration technology.
  • Quantity and total — whole pallets only, approval above 200 pallets and for colour-change ink on small Slim cans, and the can total the order states against the total computed.

Any Orora beverage can order that touches these rules is in scope — not only the scenarios pictured here. Not covered: prices, other Orora products and freight. The can formats, end sizes, end types, tabs and decoration technologies come from the public Orora brochure; the cans per pallet, minimum runs and approval limits are modeled and are not published by Orora (see the note at the foot of the page).

The AI sales agentBuilds the can 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
Options that exist for only some formatsOnly with care
Each can format takes its own end size and fill volume, so a rule keyed to the format has to hold both for every one of the twelve.		The format, fill and end size are fields of one request and are checked together on every call.
Options that exclude each otherOnly with care
The resealable and full aperture ends are limited to the 52 mm end and special tabs are excluded from the resealable end, so each exclusion needs its own rule.		Every option is a field of one request, so each exclusion is checked on every call.
A minimum that depends on another choiceOnly with care
The minimum run is 5 pallets on standard print, 10 on special effects and 5 per design on Flex FX, so a rule has to compute it from the decoration and the designs.		The decoration, designs and pallets are checked together on every call.
A stated total that must match the computed totalOnly with workarounds
The can total is pallets times the cans on a pallet for that format, 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 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.

    
1Standard_360 decoration needs at least 5 pallets for this order; the order has 4.
2The order has 0 pallets; cans are ordered in whole pallets, one pallet at least.
3The customer fills 330 mL; CLASSIC_375 holds 375 mL.
4A 52 mm end does not fit SLIM_250; the brochure lists a different end size for this can.
5A 50 mm end does not fit CLASSIC_375; the brochure lists a different end size for this can.
6The XO_Resealable end is modeled for the 52 mm (202) end only; this order uses a 50 mm end.
7The Full_Aperture end is modeled for the 52 mm (202) end only; this order uses a 50 mm end.
8A Coloured tab is not modeled for the XO resealable end.
9The brochure lists Flex FX at up to 16 designs; the order has 17.
10Flex_FX decoration needs at least 20 pallets for this order; the order has 15.
113 designs are ordered with Standard_360 decoration; more than one design per run needs Flex FX.
12Tactile_FX decoration needs at least 10 pallets for this order; the order has 9.
13Emboss_FX decoration needs at least 10 pallets for this order; the order has 3.
14Thermochromic_FX on SLIM_150 has a small print area; a person should confirm the artwork.
15Ultraviolet_FX on SLIM_300 has a small print area; a person should confirm the artwork.
16201 pallets is above 200; a person should confirm plant capacity and lead time.
17The order states 100,000 cans in total; 20 pallets of 10,752 cans each give 215,040.
18Photochromic_FX decoration needs at least 10 pallets for this order; the order has 4.
19Photochromic_FX on SLIM_150 has a small print area; a person should confirm the artwork.
20The customer fills 500 mL; CLASSIC_375 holds 375 mL.
212 designs are ordered with Standard_360 decoration; more than one design per run needs Flex FX.
22Standard_360 decoration needs at least 5 pallets for this order; the order has 3.
23The order states 1 cans in total; 3 pallets of 10,752 cans each give 32,256.
24The customer fills 355 mL; CLASSIC_330 holds 330 mL.
25A 50 mm end does not fit CLASSIC_330; the brochure lists a different end size for this can.
26The brochure lists Flex FX at up to 16 designs; the order has 20.
27Photochromic_FX decoration needs at least 10 pallets for this order; the order has 6.

  

  

    
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. End type, tab and designs: A Slim 250 mL order with an XO resealable end, a coloured tab and three designs on standard print

    
The agent's request. Please quote 30 pallets of the Slim 250 mL can filled with 250 mL, with an XO resealable end (50 mm), a coloured tab, standard 360 degree printing and 3 designs, for a total of 432,000 cans. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Resealable and full aperture ends need a 52 mm end. The XO_Resealable end is modeled for the 52 mm (202) end only; this order uses a 50 mm end.
    • 
      
  • Special tab on a resealable end. A Coloured tab is not modeled for the XO resealable end.
    • 
      
  • Several designs without Flex FX. 3 designs are ordered with Standard_360 decoration; more than one design per run needs Flex FX.
    • 
    

    
2. End type, tab and designs: The agent resubmits after fixing the errors

    
The agent's request. Please quote 30 pallets of the Slim 250 mL can filled with 250 mL, with an easy-open end (50 mm), a standard tab, standard 360 degree printing and 1 design, for a total of 432,000 cans. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Order summary. 30 pallets of SLIM_250, 14,400 cans per pallet, 432,000 cans in total.
    • 
    

    
3. Fill volume and end size: A Classic 330 mL order filled with 355 mL on a 50 mm end

    
The agent's request. Please quote 24 pallets of the Classic 330 mL can filled with 355 mL, with an easy-open end (50 mm), a standard tab, standard 360 degree printing and 1 design, for a total of 290,304 cans. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Fill volume does not match the can. The customer fills 355 mL; CLASSIC_330 holds 330 mL.
    • 
      
  • End size does not fit the can. A 50 mm end does not fit CLASSIC_330; the brochure lists a different end size for this can.
    • 
    

    
4. Fill volume and end size: The agent resubmits after fixing the errors

    
The agent's request. Please quote 24 pallets of the Classic 355 mL can filled with 355 mL, with an easy-open end (52 mm), a standard tab, standard 360 degree printing and 1 design, for a total of 274,176 cans. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Order summary. 24 pallets of CLASSIC_355, 11,424 cans per pallet, 274,176 cans in total.
    • 
    

    
5. Flex FX designs and minimum run: A Flex FX order with twenty designs on 60 pallets

    
The agent's request. Please quote 60 pallets of the Classic 375 mL can filled with 375 mL, with an easy-open end (52 mm), a standard tab, Flex FX and 20 designs, for a total of 645,120 cans. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • More than 16 designs on Flex FX. The brochure lists Flex FX at up to 16 designs; the order has 20.
    • 
      
  • Below the minimum run. Flex_FX decoration needs at least 100 pallets for this order; the order has 60.
    • 
    

    
6. Flex FX designs and minimum run: The agent resubmits after fixing the errors

    
The agent's request. Please quote 60 pallets of the Classic 375 mL can filled with 375 mL, with an easy-open end (52 mm), a standard tab, Flex FX and 12 designs, for a total of 645,120 cans. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Order summary. 60 pallets of CLASSIC_375, 10,752 cans per pallet, 645,120 cans 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 beverage cans: a can format and fill volume, an end size and end type, a tab, and more. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the Orora Beverage Cans range and returns a verdict, every reason and the can total it computed.
    2. 
      
  2. I will start with the first request by the agent.
    3. 
      
  3. The agent says. Please quote 30 pallets of the Slim 250 mL can filled with 250 mL, with an XO resealable end (50 mm). 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 three findings, all in one answer. The XO_Resealable end is modeled for the 52 mm (202) end only; this order uses a 50 mm end. A Coloured tab is not modeled for the XO resealable end. 3 designs are ordered with Standard_360 decoration; more than one design per run needs Flex FX.
    6. 
      
  6. The validator returned three 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 options that exclude each other, a rules-based CPQ is error-prone for this.
    8. 
      
  8. Now the agent resubmits.
    9. 
      
  9. The corrected request. It changes the end type to easy-open, the tab to standard and the designs to 1.
    10. 
      
  10. It calls the validator again.
    11. 
      
  11. This time the validator clears it.
    12. 
      
  12. All three earlier findings are cleared.
    13. 
      
  13. Three 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.