Predictika · Container shipping bookings

An AI sales agent's container booking request, checked before it goes out

An AI sales agent prepares a booking request for full containers: a container type, a number of containers, and the cargo weight, volume and height per container. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the published container specifications and returns a verdict, every reason and the figures it computed.

Scope: Hapag-Lloyd FCL container bookings: cargo weight, volume and height against the published specifications of six container types. See everything this validator covers

This validator models full-container bookings on six Hapag-Lloyd container types, not just the scenarios shown below.

  • Container types — 20' standard, 40' standard, 40' standard high cube, 20' reefer, 40' reefer high cube and 40' flatrack high cube / platform, with the figures from the public Hapag-Lloyd container pages.
  • Cargo against the container — cargo weight per container against the maximum payload, cargo volume against the capacity, and cargo height against the door opening (flatracks publish no capacity or door, so only weight is checked).
  • Stated figures — the total cargo weight the request states must equal the number of containers times the weight per container; a load at exactly the full published payload goes to a person to confirm, because Hapag-Lloyd says its published figures are examples and vary by manufacturer.

Any container booking that touches these rules is in scope — not only the scenarios pictured here. Not covered: freight rates, surcharges, lane and schedule terms, volume commitments (this validator has no price list, so it produces no price build-up or line-item quote), other container types (open top, hardtop, tank, 45', special equipment), dangerous goods, reefer temperature settings and cargo stowage.

The AI sales agentBuilds the booking request from the conversation and says it is ready.
Predictika validatorChecks every rule at once and returns the verdict, each finding, and the computed figures.

7 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
Cargo weight, volume and height all checked against the container typeOnly with care
Each limit depends on the container type chosen, so a rule set needs a table per type and a rule per limit, and the table has to be kept in step with the product list.		The container type, its limits and the cargo figures are in one request and every limit is checked on every call.
Cargo that fits a high cube door but not a standard doorOnly with workarounds
The door height differs by type, so the same cargo passes on one type and fails on another; a rep usually finds out when the booking is reviewed.		The cargo height is compared with the door opening of whichever type is chosen, so switching the type changes the result at once.
Stated total weight that must equal containers times weight per containerOnly with care
A stated figure is a free-entry field; a rule can compare it with a computed total, but someone has to write and maintain that comparison.		The stated figure travels with the booking and is compared with the computed figure on every call.
Every problem returned in one answerOnly with workarounds
Rules tend to fire as the booking 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 cargo weighs 28,131 kg per container; the published maximum payload for this container type is 28,130 kg.
2The cargo weighs 29,000 kg per container; the published maximum payload for this container type is 28,750 kg.
3The cargo weighs 28,601 kg per container; the published maximum payload for this container type is 28,600 kg.
4The cargo weight of 28,600 kg equals the full published payload; Hapag-Lloyd states its published specifications are examples and containers vary by manufacturer, so a person should confirm the actual container.
5The cargo takes 33,201 litres per container; the published capacity of this container type is 33,200 litres.
6The cargo is 2,500 mm high; the door opening of this container type is 2,292 mm, so it cannot be loaded through the doors.
7The cargo is 2,598 mm high; the door opening of this container type is 2,597 mm, so it cannot be loaded through the doors.
8The cargo takes 30,000 litres per container; the published capacity of this container type is 28,100 litres.
9The cargo is 2,600 mm high; the door opening of this container type is 2,557 mm, so it cannot be loaded through the doors.
10The cargo weighs 49,101 kg per container; the published maximum payload for this container type is 49,100 kg.
11The quote states 120,000 kg of cargo in total; the containers and cargo weight give 144,000 kg.
12A booking needs at least 1 container; the quote has 0.
13The cargo weighs 30,000 kg per container; the published maximum payload for this container type is 28,130 kg.
14The cargo takes 40,000 litres per container; the published capacity of this container type is 33,200 litres.
15The cargo weight of 49,100 kg equals the full published payload; Hapag-Lloyd states its published specifications are examples and containers vary by manufacturer, so a person should confirm the actual container.

  

  

    
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. Payload and container type: A booking with cargo heavier than the container allows

    
The agent's request. Please book 8 40' standards carrying 29,500 kg of cargo each, 62,000 litres in volume, 2,200 mm high. The request states 236,000 kg of cargo in total. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Cargo over maximum payload. The cargo weighs 29,500 kg per container; the published maximum payload for this container type is 28,750 kg.
    • 
    

    
2. Payload and container type: The agent resubmits after fixing the errors

    
The agent's request. Please book 8 40' standard high cubes carrying 28,000 kg of cargo each, 62,000 litres in volume, 2,200 mm high. The request states 224,000 kg of cargo in total. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Booking summary. 8 container(s) carrying 28,000 kg per container, 224,000 kg in total.
    • 
    

    
3. Cargo height and door: A booking with cargo too tall for the doors

    
The agent's request. Please book 10 40' standards carrying 26,000 kg of cargo each, 66,000 litres in volume, 2,450 mm high. The request states 260,000 kg of cargo in total. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Cargo taller than door opening. The cargo is 2,450 mm high; the door opening of this container type is 2,292 mm, so it cannot be loaded through the doors.
    • 
    

    
4. Cargo height and door: The agent resubmits after fixing the errors

    
The agent's request. Please book 10 40' standard high cubes carrying 26,000 kg of cargo each, 66,000 litres in volume, 2,450 mm high. The request states 260,000 kg of cargo in total. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Booking summary. 10 container(s) carrying 26,000 kg per container, 260,000 kg in total.
    • 
    

    
5. Reefer volume and stated total: A reefer booking with too much cargo volume and a wrong total

    
The agent's request. Please book 12 20' reefers carrying 26,500 kg of cargo each, 29,000 litres in volume, 2,200 mm high. The request states 300,000 kg of cargo in total. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Cargo over capacity. The cargo takes 29,000 litres per container; the published capacity of this container type is 28,100 litres.
    • 
      
  • Stated total weight does not match. The quote states 300,000 kg of cargo in total; the containers and cargo weight give 318,000 kg.
    • 
    

    
6. Reefer volume and stated total: The agent resubmits after fixing the errors

    
The agent's request. Please book 12 20' reefers carrying 26,500 kg of cargo each, 28,000 litres in volume, 2,200 mm high. The request states 300,000 kg of cargo in total. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Stated total weight does not match. The quote states 300,000 kg of cargo in total; the containers and cargo weight give 318,000 kg.
    • 
    

    
7. Reefer volume and stated total: The agent resubmits after fixing the errors

    
The agent's request. Please book 12 20' reefers carrying 26,500 kg of cargo each, 28,000 litres in volume, 2,200 mm high. The request states 318,000 kg of cargo in total. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Booking summary. 12 container(s) carrying 26,500 kg per container, 318,000 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 a booking request for full containers: a container type, a number of containers, the cargo weight, and more. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the published container specifications and returns a verdict, every reason and the figures it computed.
    2. 
      
  2. I will start with the first request by the agent.
    3. 
      
  3. The agent says. Please book 12 20' reefers carrying 26,500 kg of cargo each, 29,000 litres in volume, 2,200 mm high. The request states 300,000 kg of cargo in total. 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 cargo takes 29,000 litres per container; the published capacity of this container type is 28,100 litres. The quote states 300,000 kg of cargo in total; the containers and cargo weight give 318,000 kg.
    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 and against the stated total.
    7. 
      
  7. What this shows: for stated total weight that must equal containers times weight per container, a rules-based CPQ is error-prone for this.
    8. 
      
  8. Now the agent resubmits.
    9. 
      
  9. The corrected request. It changes the cargo volume per container to 28,000 litres.
    10. 
      
  10. It calls the validator again.
    11. 
      
  11. Still blocked, with one finding, all in one answer. The quote states 300,000 kg of cargo in total; the containers and cargo weight give 318,000 kg.
    12. 
      
  12. One finding cleared. Still blocking: stated total weight does not match.
    13. 
      
  13. The agent resubmits once more.
    14. 
      
  14. The revised request. It changes the stated total cargo to 318,000 kg.
    15. 
      
  15. It calls the validator once more.
    16. 
      
  16. This time the validator clears it.
    17. 
      
  17. The earlier finding is cleared.
    18. 
      
  18. Two problems caught in one call, fixed in two passes, 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.
    19.