Predictika · Robot cell quotes

An AI sales agent's cobot cell quote, checked before it goes out

An AI sales agent prepares a quote for a Universal Robots cobot cell, with a robot model, a tool load, the reach and repeatability the application needs, an ingress protection rating, a teach pendant, palletizing and force control options, and a rated payload for the robot. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the e-Series and UR Series robots and returns a verdict and every reason.

Scope: Universal Robots cobot cell quotes: the e-Series UR3e, UR7e, UR12e and UR16e and the UR Series UR15, UR20 and UR30, with payload, reach, repeatability, protection rating, pendant, palletizing, force control and tooling checks. See everything this validator covers

This validator models Universal Robots cobot cell quotes, not just the scenarios shown below.

  • Robot limits — the rated and extended payload, the reach and the repeatability of each of the seven models, checked against the tool load and application the quote needs, and the IP65 rating.
  • Cell contents — a teach pendant on the quote, palletizing, which needs a robot with at least 1,300 mm of reach, force control, which needs an external sensor on the UR Series, and whether the end-of-arm tool is UR+ validated.
  • Stated payload — the rated payload the quote states against the rated payload of the quoted robot.

Any Universal Robots cobot cell quote that touches these rules is in scope — not only the scenarios pictured here. Not covered: newer variants (UR8 Long, UR18, g-Series and the long-reach variants), control box and software options, and prices; the validator models no prices. The model names, payloads and reach follow the public Universal Robots product pages; the e-Series repeatability figures, the compatibility rules and every approval rule are modeled from comparable cobot products and are not published by Universal Robots.

The AI sales agentBuilds the cobot cell quote 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
Limits that depend on the robot modelOnly with care
Payload, reach and repeatability differ on each of the seven models, so every limit needs its own list by model.		The model and the application needs are in one request and every limit is checked together on every call.
A load that is allowed only with approvalOnly with workarounds
The extended payload is allowed on the UR Series at reduced speed, so the rule has to separate a block from an approval.		The tool load is compared with both the rated and the extended payload on every call and returns a block, an approval or nothing.
An option that depends on the modelOnly with care
Palletizing needs long reach and force control needs an external sensor on the UR Series, so each option needs a rule that looks at the model.		The model and the options are in the same request, so every dependency is checked on every call.
A stated payload that must match the robotOnly with workarounds
The rated payload depends on the model, so a separate rule is needed to look it up and compare it with the stated figure.		The stated payload is a field of the request and is compared with the rated payload of the quoted robot 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 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 gripper and part weigh 80 (x0.1 kg) at the flange; this robot is rated for at most 75 (x0.1 kg).
2The gripper and part weigh 161 (x0.1 kg) at the flange; this robot is rated for at most 160 (x0.1 kg).
3The tool load is above the rated payload and within the extended payload; a person should confirm the reduced-speed duty cycle.
4The gripper and part weigh 180 (x0.1 kg) at the flange; this robot is rated for at most 175 (x0.1 kg).
5The cell needs 1,800 mm of reach; this robot reaches 1,750 mm.
6The cell needs 1,400 mm of reach; this robot reaches 1,300 mm.
7The application needs repeatability of 60 micrometres; this robot is rated for 100.
8The cell needs ingress protection IP67; the robot is rated IP65.
9The quote has no teach pendant or smart panel to program and start the robot.
10A palletizing cell needs a robot with at least 1,300 mm of reach; this robot reaches 900 mm.
11The quote states a rated payload of 160 (x0.1 kg); the quoted robot is rated 125.
12Force control is needed and this robot has no embedded force-torque sensor; a person should confirm an external sensor is quoted.
13The end-of-arm tool is not a UR+ validated product; a person should confirm integration.
14The gripper and part weigh 60 (x0.1 kg) at the flange; this robot is rated for at most 30 (x0.1 kg).
15The cell needs 700 mm of reach; this robot reaches 500 mm.

  

  

    
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, reach, pendant and stated payload: A UR7e cell with a 9 kg tool, 1,000 mm of reach, no teach pendant and a stated payload of 10 kg

    
The agent's request. Please quote 2 Universal Robots UR7e robots with a 9 kg tool load, 1,000 mm of reach needed, repeatability of 100 micrometers and ingress protection IP54, and a rated payload of 10 kg. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Tool payload over the extended rating. The gripper and part weigh 90 (x0.1 kg) at the flange; this robot is rated for at most 75 (x0.1 kg).
    • 
      
  • Reach too short. The cell needs 1,000 mm of reach; this robot reaches 850 mm.
    • 
      
  • No programming pendant on the quote. The quote has no teach pendant or smart panel to program and start the robot.
    • 
      
  • Stated payload does not match. The quote states a rated payload of 100 (x0.1 kg); the quoted robot is rated 75.
    • 
    

    
2. Model, reach, pendant and stated payload: The agent resubmits after fixing the errors

    
The agent's request. Please quote 2 Universal Robots UR12e robots with a 9 kg tool load, 1,000 mm of reach needed, repeatability of 100 micrometers and ingress protection IP54, with a teach pendant, and a rated payload of 12.5 kg. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Robot summary. 2 robot(s), rated payload 125 (x0.1 kg), reach 1,300 mm, repeatability 50 micrometres.
    • 
    

    
3. Palletizing and repeatability: A UR16e palletizing cell that needs repeatability of 30 micrometers

    
The agent's request. Please quote 1 Universal Robots UR16e robot with a 10 kg tool load, 900 mm of reach needed, repeatability of 30 micrometers and ingress protection IP54, with a teach pendant and palletizing, and a rated payload of 16 kg. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Repeatability finer than the robot. The application needs repeatability of 30 micrometres; this robot is rated for 50.
    • 
      
  • Palletizing needs long reach. A palletizing cell needs a robot with at least 1,300 mm of reach; this robot reaches 900 mm.
    • 
    

    
4. Palletizing and repeatability: The agent resubmits after fixing the errors

    
The agent's request. Please quote 1 Universal Robots UR15 robot with a 10 kg tool load, 900 mm of reach needed, repeatability of 50 micrometers and ingress protection IP54, with a teach pendant and palletizing, and a rated payload of 15 kg. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Robot summary. 1 robot(s), rated payload 150 (x0.1 kg), reach 1,300 mm, repeatability 50 micrometres.
    • 
    

    
5. Payload, protection and force control: A UR30 washdown cell with a 40 kg tool, IP67 and force control

    
The agent's request. Please quote 1 Universal Robots UR30 robot with a 40 kg tool load, 1,200 mm of reach needed, repeatability of 100 micrometers and ingress protection IP67, with a teach pendant and force control, and a rated payload of 30 kg. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Tool payload over the extended rating. The gripper and part weigh 400 (x0.1 kg) at the flange; this robot is rated for at most 350 (x0.1 kg).
    • 
      
  • Ingress protection above IP65. The cell needs ingress protection IP67; the robot is rated IP65.
    • 
      
  • Force control needs an external sensor. Force control is needed and this robot has no embedded force-torque sensor; a person should confirm an external sensor is quoted.
    • 
    

    
6. Payload, protection and force control: The agent resubmits after fixing the errors

    
The agent's request. Please quote 1 Universal Robots UR30 robot with a 30 kg tool load, 1,200 mm of reach needed, repeatability of 100 micrometers and ingress protection IP65, with a teach pendant and force control, and a rated payload of 30 kg. It is ready to send.

    
Verdict: Needs approval.

    
    
      
  • Force control needs an external sensor. Force control is needed and this robot has no embedded force-torque sensor; a person should confirm an external sensor is quoted.
    • 
      
  • Robot summary. 1 robot(s), rated payload 300 (x0.1 kg), reach 1,300 mm, repeatability 100 micrometres.
    • 
    

  

  

    
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 quote for a Universal Robots cobot cell, with a robot model, a tool load. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the e-Series and UR Series robots and returns a verdict and every reason.
    2. 
      
  2. I will start with the first request by the agent.
    3. 
      
  3. The agent says. Please quote 2 Universal Robots UR7e robots with a 9 kg tool load, 1,000 mm of reach needed, repeatability of 100 micrometers and ingress protection IP54. 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 four findings, all in one answer. The gripper and part weigh 90 (x0.1 kg) at the flange; this robot is rated for at most 75 (x0.1 kg). The cell needs 1,000 mm of reach; this robot reaches 850 mm. The quote has no teach pendant or smart panel to program and start the robot. The other one is on screen.
    6. 
      
  6. The validator returned four 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 stated payload that must match the robot, a rules-based CPQ can do this only with workarounds.
    8. 
      
  8. Now the agent resubmits.
    9. 
      
  9. The corrected request. It changes the robot model to UR12e, the teach pendant on the quote to yes and the rated payload stated on the quote to 12.5.
    10. 
      
  10. It calls the validator again.
    11. 
      
  11. This time the validator clears it.
    12. 
      
  12. All four earlier findings are cleared.
    13. 
      
  13. Four problems 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.