Predictika · Industrial gear units

An AI sales agent's gear unit selection, checked before it goes out

An AI sales agent selects a Flender type H2 two-stage helical gear unit for a customer's machine: size, nominal ratio, motor speed, cooling and site conditions. Before it goes out, the agent sends the selection to Predictika, which checks it against a declarative model of the rating and thermal tables and selection rules in the public Flender catalog MD 20.1 and returns a verdict, every reason and the figures it computed.

Scope: Flender type H2 helical gear units, sizes 4 to 12, for motor speeds of 1000, 1200, 1500 and 1800 rpm: nominal power ratings, service factors, peak torque and thermal capacity with the selected cooling. See everything this validator covers

This validator models the selection procedure of Flender catalog MD 20.1 (edition 2018) for the type H2 gear unit, not just the scenarios shown below.

  • Nominal power rating — the rating of each size 4 to 12 for the 14 nominal ratios (6.3 to 28) at each of the four motor speeds, from the catalog rating table, checked against the driven power times the driven machine factor and the prime mover factor.
  • Driven machine and duty — the catalog's machine factor for 77 driven machines (waste water, pumps, excavators, chemical, metal working, rolls, conveyors, compressors, food, paper, cable railways, cement) for three operating periods under load, and its thermal factor.
  • Peak torque — the peak torque factor from load peaks per hour and steady or alternating direction, checked against the nominal rating.
  • Thermal capacity — the catalog thermal capacity for no auxiliary cooling, fan, cooling coil, or fan with cooling coil, corrected for ambient temperature and PAO or mineral oil, compared with the capacity the machine needs.
  • Items a person confirms — cells the catalog marks as on request, machines or operating periods with no catalog factor, ambient temperatures between the tabulated steps, altitude above 1000 m, installation outside a large hall, an on-load factor below 100%, and a gear unit more than 3.33 times the load.

Any type H2 selection that touches these rules is in scope — not only the scenarios pictured here. Not covered: prices (Flender publishes none), sizes 13 to 28, the other gear unit types (H1, H3, H4, B2, B3, B4), crane applications, variable power ratings, upright or vertical mounting, forced lubrication, ATEX, shaft loads, bearing life and converter operation.

The AI sales agentSelects the gear unit 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
A nominal rating that depends on size, nominal ratio and motor speed together (a three-way table with gaps)Only with care
Each cell becomes a lookup row or a product rule; with 9 sizes, 14 ratios and 4 speeds that is hundreds of values, and the empty cells need their own handling.		The three choices travel in one request and are looked up together, including sizes the catalog does not offer for a ratio.
A required rating computed from the load, a machine factor, a prime mover factor and a peak torque factorOnly with workarounds
The factors come from several tables and the comparison is arithmetic over their product, which a rules-based CPQ usually does with custom code before it can compare the result with the table value.		Every factor is looked up and the required figures are computed and compared on every call.
A thermal capacity that depends on cooling, ambient temperature and oil, compared with a need that depends on the machineOnly with workarounds
The usable capacity is a table value times a temperature factor chosen by oil type, compared with a figure that depends on the driven machine; each step is a separate lookup or script.		The capacity with the selected cooling and the capacity needed are computed together, so an under-cooled unit is caught.
Every problem returned in one answerOnly with workarounds
Rules tend to fire as the selection 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 catalog rates type H2 for motor speeds of 1000, 1200, 1500 and 1800 rpm; the quote has 1400 rpm.
2This validator covers type H2 gear unit sizes 4 to 12; the quote has size 3.
3This validator covers type H2 gear unit sizes 4 to 12; the quote has size 13.
4The catalog lists no type H2 size 6 for this nominal ratio and speed.
5The catalog lists no type H2 size 7 for this nominal ratio and speed.
6The nominal power rating of 141 kW is below the 165 kW required (driven power times the machine and prime mover factors).
7The thermal capacity with the selected cooling is 87.6 kW; 93.5 kW is needed, so more cooling is required.
8The thermal capacity with the selected cooling is 71.7 kW; 90 kW is needed, so more cooling is required.
9The nominal power rating of 199 kW is below the 294.5 kW needed for the peak torque.
10The thermal capacity with the selected cooling is 52.3 kW; 93.5 kW is needed, so more cooling is required.
11The thermal capacity with the selected cooling is 74 kW; 93.5 kW is needed, so more cooling is required.
12The quote states a nominal power rating of 199 kW; the catalog rating for this size, ratio and speed is 141 kW.
13The quote states a nominal power rating of 110 kW; the catalog rating for this size, ratio and speed is 141 kW.
14The nominal rating of 235 kW is more than 3.33 times the driven power of 70 kW; the catalog asks for a consultation with Flender in that case.
15The catalog gives no service factor for this machine and operating period; Flender must be consulted.
16The catalog gives thermal factors only for ambient temperatures from 10 to 50 C in steps of 5 C; 22 C needs Flender.

  

  

    
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. Hot site: An apron conveyor in a 40 C plant, no auxiliary cooling

    
The agent's request. Please quote a Flender type H2 gear unit, size 7, nominal ratio 16, with a 1500 rpm motor, for apron conveyors. The driven power is 110 kW, driven by an electric motor, under load for more than 10 hours a day, with 6 to 30 load peaks per hour in a steady direction and a peak torque of 900 Nm. The site is at 40 C and 500 m, installed in a large hall, with no auxiliary cooling and mineral oil, at an on-load factor of 100%. The nominal power rating of the unit is 199 kW. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Thermal capacity too low for the selected cooling. The thermal capacity with the selected cooling is 52.3 kW; 93.5 kW is needed, so more cooling is required.
    • 
    

    
2. Hot site: The agent resubmits after fixing the errors

    
The agent's request. Please quote a Flender type H2 gear unit, size 7, nominal ratio 16, with a 1500 rpm motor, for apron conveyors. The driven power is 110 kW, driven by an electric motor, under load for more than 10 hours a day, with 6 to 30 load peaks per hour in a steady direction and a peak torque of 900 Nm. The site is at 40 C and 500 m, installed in a large hall, with fan cooling and mineral oil, at an on-load factor of 100%. The nominal power rating of the unit is 199 kW. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Selection summary. Nominal rating 199 kW against 165 kW required; thermal capacity 139.9 kW against 93.5 kW needed.
    • 
    

    
3. Peak torque: The same conveyor with heavy alternating load peaks

    
The agent's request. Please quote a Flender type H2 gear unit, size 7, nominal ratio 16, with a 1500 rpm motor, for apron conveyors. The driven power is 110 kW, driven by an electric motor, under load for more than 10 hours a day, with more than 100 load peaks per hour in an alternating direction and a peak torque of 1500 Nm. The site is at 20 C and 500 m, installed in a large hall, with no auxiliary cooling and synthetic PAO oil, at an on-load factor of 100%. The nominal power rating of the unit is 199 kW. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Rating too low for the peak torque. The nominal power rating of 199 kW is below the 294.5 kW needed for the peak torque.
    • 
    

    
4. Peak torque: The agent resubmits after fixing the errors

    
The agent's request. Please quote a Flender type H2 gear unit, size 9, nominal ratio 16, with a 1500 rpm motor, for apron conveyors. The driven power is 110 kW, driven by an electric motor, under load for more than 10 hours a day, with more than 100 load peaks per hour in an alternating direction and a peak torque of 1500 Nm. The site is at 20 C and 500 m, installed in a large hall, with no auxiliary cooling and synthetic PAO oil, at an on-load factor of 100%. The nominal power rating of the unit is 331 kW. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Selection summary. Nominal rating 331 kW against 165 kW required; thermal capacity 123 kW against 93.5 kW needed.
    • 
    

    
5. Ratio and size: A pump drive at the highest nominal ratio

    
The agent's request. Please quote a Flender type H2 gear unit, size 7, nominal ratio 28, with a 1500 rpm motor, for centrifugal pumps. The driven power is 70 kW, driven by an electric motor, under load for more than 10 hours a day, with 1 to 5 load peaks per hour in a steady direction and a peak torque of 500 Nm. The site is at 20 C and 500 m, installed in a large hall, with no auxiliary cooling and synthetic PAO oil, at an on-load factor of 100%. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Size not offered for this ratio. The catalog lists no type H2 size 7 for this nominal ratio and speed.
    • 
    

    
6. Ratio and size: The agent resubmits after fixing the errors

    
The agent's request. Please quote a Flender type H2 gear unit, size 10, nominal ratio 28, with a 1500 rpm motor, for centrifugal pumps. The driven power is 70 kW, driven by an electric motor, under load for more than 10 hours a day, with 1 to 5 load peaks per hour in a steady direction and a peak torque of 500 Nm. The site is at 20 C and 500 m, installed in a large hall, with no auxiliary cooling and synthetic PAO oil, at an on-load factor of 100%. The nominal power rating of the unit is 235 kW. It is ready to send.

    
Verdict: Needs approval.

    
    
      
  • Gear unit much larger than the load. The nominal rating of 235 kW is more than 3.33 times the driven power of 70 kW; the catalog asks for a consultation with Flender in that case.
    • 
      
  • Selection summary. Nominal rating 235 kW against 91 kW required; thermal capacity 111 kW against 70 kW needed.
    • 
    

    
7. Ratio and size: The agent resubmits after fixing the errors

    
The agent's request. Please quote a Flender type H2 gear unit, size 8, nominal ratio 28, with a 1500 rpm motor, for centrifugal pumps. The driven power is 70 kW, driven by an electric motor, under load for more than 10 hours a day, with 1 to 5 load peaks per hour in a steady direction and a peak torque of 500 Nm. The site is at 20 C and 500 m, installed in a large hall, with no auxiliary cooling and synthetic PAO oil, at an on-load factor of 100%. The nominal power rating of the unit is 141 kW. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Selection summary. Nominal rating 141 kW against 91 kW required; thermal capacity 91.6 kW against 70 kW needed.
    • 
    

  

  

    
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 selects a Flender type H2 two-stage helical gear unit for a customer's machine: size, nominal ratio, motor speed, and more. Before it goes out, the agent sends the selection to Predictika, which checks it against a declarative model of the rating and thermal tables and selection rules in the public Flender catalog MD 20.1 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 quote a Flender type H2 gear unit, size 7, nominal ratio 28, with a 1500 rpm motor, for centrifugal pumps. The driven power is 70 kW, driven by an electric motor, under load for more than 10 hours a day. The rest of the request is on screen.
    4. 
      
  4. Before it goes out, the agent calls the validator.
    5. 
      
  5. Blocked, with one finding, all in one answer. The catalog lists no type H2 size 7 for this nominal ratio and speed.
    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 a nominal rating that depends on size, nominal ratio and motor speed together (a three-way table with gaps), a rules-based CPQ is error-prone for this.
    8. 
      
  8. Now the agent resubmits.
    9. 
      
  9. The corrected request. It changes the gear unit to Type H2, size 10. It sets the stated nominal rating to 235 kW.
    10. 
      
  10. It calls the validator again.
    11. 
      
  11. Not blocked, but held for a person.
    12. 
      
  12. The one blocking finding is cleared; one approval remains for a person: gear unit much larger than the load.
    13. 
      
  13. The agent resubmits once more.
    14. 
      
  14. The revised request. It changes the gear unit to Type H2, size 8 and the stated nominal rating to 141 kW.
    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. One problem 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.