Predictika · Container bookings

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

An AI sales agent prepares an ocean container booking: the container type, Spot or contract, how many containers, the cargo weight, a temperature setting, the cargo class, free days, the days to departure and the total weight it states. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the booking rules and returns a verdict, every reason and the figures it computed.

Scope: Maersk Spot and contract ocean container bookings: dry, high cube, 45 foot, reefer, open top and flat rack containers, weight, temperature, cargo class, free time, quantity, timing and the stated total weight. See everything this validator covers

This validator models container bookings of this kind, not just the scenarios shown below.

  • Container and weight — the nine container types and the weight each takes, and the stated total weight compared with the containers.
  • Reefers and cargo class — the temperature a reefer needs and the range accepted, dangerous goods, explosives that cannot be booked online, and over-dimension cargo that needs an open top or flat rack.
  • Product, quantity and timing — the annual commitment a contract needs, free days for Spot and contract, the largest booking, the booking cut-off before departure, and large Spot bookings that need space confirmed.

Any container booking that touches these rules is in scope — not only the scenarios pictured here. The container types and the Spot and contract products follow Maersk's public container FAQ and Spot terms; the payload limits, the reefer range, the dangerous goods and over-dimension handling, the commitment, the free-time limits, the booking size limit and the cut-off are modeled from comparable carrier practice and not published by Maersk. No freight rates are modeled. Not covered: tank containers, lane rates, surcharges and inland legs.

The AI sales agentBuilds the container booking 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 weight limit that depends on the container typeOnly with care
Each of the nine container types takes a different weight, so the weight needs a rule per type.
The container and weight are fields of one booking and are checked together.
Reefer rulesOnly with care
A reefer needs a temperature within a range and no other container may have one, so three rules depend on the container.
The container and temperature are in one request and all three conditions are evaluated on every call.
Limits that depend on the product and the timingOnly with workarounds
The free days, the annual commitment, the booking size and the cut-off before departure depend on Spot or contract and on the days to departure.
The product, quantity, free days and days to departure are in one request and all the conditions are evaluated on every call.
A stated total weightOnly with workarounds
The total weight has to be computed from the containers and compared with the total stated.
The total is computed on every call and compared with the total stated in the booking.
Every problem returned in one answerOnly with workarounds
Rules tend to fire as a step is saved or 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 booking; approval findings need a person to sign off; everything else is computed.

1The cargo weighs 27,500 kg per container; this container type takes at most 26,700 kg.
2A reefer booking needs a temperature setting.
3The reefer set point is -40 C; the range accepted is -35 C to +30 C.
4A temperature setting applies to reefer containers only; this is not a reefer.
5Class 1 explosives cannot be booked online.
6Dangerous goods need a declaration review before the booking is confirmed.
7Over-dimension cargo needs an open top or flat rack container.
8Over-dimension cargo needs a special-cargo review.
9A contract needs an annual commitment of at least 50 FEU; the booking has 30.
1015 free days requested; this product allows at most 14.
11A spot booking of 20 or more containers needs confirmation of space.
12101 containers on one booking; the limit is 100.
13The vessel departs in 1 day(s); bookings close 2 days before departure.
14The booking states 90,000 kg in total; the containers give 80,000 kg.
15The cargo weighs 28,000 kg per container; this container type takes at most 27,600 kg.
16The cargo weighs 29,000 kg per container; this container type takes at most 28,100 kg.

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. Dry containers and timing: A booking of dry containers that are too heavy, with a temperature setting, over-dimension cargo, too many free days and a departure too close

The agent's request. Please book 4 40-foot dry containers on a Spot booking, 27,500 kg in each, with a temperature set to 4 degrees Celsius, carrying over-dimension cargo. The booking asks for 15 free days and the vessel departs in 1 days. The booking states a total weight of 110,000 kg. It is ready to send.

Verdict: Blocked.

  • Cargo weight over container payload. The cargo weighs 27,500 kg per container; this container type takes at most 26,700 kg.
  • Temperature set on a non-reefer. A temperature setting applies to reefer containers only; this is not a reefer.
  • Over-dimension cargo in a closed container. Over-dimension cargo needs an open top or flat rack container.
  • Free time over the product maximum. 15 free days requested; this product allows at most 14.
  • Booking inside the cut-off. The vessel departs in 1 day(s); bookings close 2 days before departure.

2. Dry containers and timing: The agent resubmits the cargo in reefers

The agent's request. Please book 4 40-foot reefer containers on a Spot booking, 27,500 kg in each, with a temperature set to 4 degrees Celsius. The booking asks for 7 free days and the vessel departs in 14 days. The booking states a total weight of 110,000 kg. It is ready to send.

Verdict: Allowed.

  • Booking summary. 4 container(s), 110,000 kg total, 29,000 kg container limit, 7 free days.

3. Contract size and cargo class: A contract booking with explosives, too small a commitment and too many containers

The agent's request. Please book 101 40-foot dry containers on a Contract booking, 20,000 kg in each, carrying dangerous goods class 1, under a contract with an annual commitment of 30 FEU. The booking asks for 21 free days and the vessel departs in 14 days. The booking states a total weight of 2,020,000 kg. It is ready to send.

Verdict: Blocked.

  • Explosives not bookable. Class 1 explosives cannot be booked online.
  • Contract below minimum annual commitment. A contract needs an annual commitment of at least 50 FEU; the booking has 30.
  • Too many containers on one booking. 101 containers on one booking; the limit is 100.

4. Contract size and cargo class: The agent resubmits after fixing the errors

The agent's request. Please book 40 40-foot dry containers on a Contract booking, 20,000 kg in each, under a contract with an annual commitment of 120 FEU. The booking asks for 21 free days and the vessel departs in 14 days. The booking states a total weight of 800,000 kg. It is ready to send.

Verdict: Allowed.

  • Booking summary. 40 container(s), 800,000 kg total, 26,700 kg container limit, 21 free days.

5. Reefer temperature and total: A reefer booking set too cold, with a total weight that does not match and a departure too close

The agent's request. Please book 4 40-foot reefer containers on a Spot booking, 20,000 kg in each, with a temperature set to -40 degrees Celsius. The booking asks for 7 free days and the vessel departs in 1 days. The booking states a total weight of 90,000 kg. It is ready to send.

Verdict: Blocked.

  • Reefer temperature out of range. The reefer set point is -40 C; the range accepted is -35 C to +30 C.
  • Booking inside the cut-off. The vessel departs in 1 day(s); bookings close 2 days before departure.
  • Stated total weight does not match. The booking states 90,000 kg in total; the containers give 80,000 kg.

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

The agent's request. Please book 4 40-foot reefer containers on a Spot booking, 20,000 kg in each, with a temperature set to -18 degrees Celsius. The booking asks for 7 free days and the vessel departs in 14 days. The booking states a total weight of 80,000 kg. It is ready to send.

Verdict: Allowed.

  • Booking summary. 4 container(s), 80,000 kg total, 29,000 kg container limit, 7 free days.

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 ocean container booking: the container type, Spot or contract, how many containers, and more. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the booking rules and returns a verdict, every reason and the figures it computed.
  2. I will start with the first request by the agent.
  3. The agent says. Please book 4 40-foot dry containers on a Spot booking, 27,500 kg in each, with a temperature set to 4 degrees Celsius, carrying over-dimension cargo. The booking asks for 15 free days and the vessel departs in 1 days. The rest of the request is on screen. The booking states a total weight of 110,000 kg.
  4. Before it goes out, the agent calls the validator.
  5. Blocked, with five findings, all in one answer. The cargo weighs 27,500 kg per container; this container type takes at most 26,700 kg. A temperature setting applies to reefer containers only; this is not a reefer. Over-dimension cargo needs an open top or flat rack container. The other two are on screen.
  6. The validator returned five 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 booking are checked against each other.
  7. What this shows: for a weight limit that depends on the container type, a rules-based CPQ is error-prone for this.
  8. Now the agent resubmits.
  9. The corrected request. It changes the container to 40-foot reefer, the over-dimension to no, the free days to 7 and the days to departure to 14.
  10. It calls the validator again.
  11. This time the validator clears it.
  12. All five earlier findings are cleared.
  13. Five problems caught in one call, fixed in one pass, and a clean booking. That is the value of Predictika's validator: catching mistakes before the booking goes out. Now you can try the other scenarios yourself.