Predictika · Rail and turnout quotes

An AI sales agent's rail and turnout quote, checked before it goes out

An AI sales agent prepares a quote for a rail operator: rail grade and tonnes, track type, axle load, site temperatures, turnouts with their diverging speed, Intelligent Turnout monitoring, a discount and the gross the quote states. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the grade, track and turnout rules and returns a verdict, every reason and the totals it computed.

Scope: voestalpine Railway Systems rail and turnout quotes: rail grades, Vignol and grooved track, standard, grooved and high-speed turnouts, Intelligent Turnout monitoring, discounts and the gross the quote states. See everything this validator covers

This validator models rail and turnout quotes, not just the scenarios shown below.

  • Rail grades and track — R260, 350 HT HSH, 340 Dobain HSH, 400 UHC HSH and the grooved 290 GHT-CL HSH, 340 GHT HSH and 400 GHT grades, on Vignol or grooved track, for mixed, heavy-haul, metro and tram, or high-speed use.
  • Turnouts — standard, grooved and high-speed turnouts with their diverging speed, the track type and rail grade they go with, and Intelligent Turnout monitoring.
  • Totals and approvals — minimum rail lot, discounts, custom profiles, very large rail orders, and the gross the quote states against the gross computed from the items.

Any rail and turnout quote that touches these rules is in scope — not only the scenarios pictured here. Not covered: fastenings, bearers, point machines and services. The product and grade names, the 400 UHC HSH temperature range of -40 to +50 degrees C, the 40 t axle load of heavy haul and of turnouts, and the existence of high-speed turnouts and Intelligent Turnout monitoring follow voestalpine's public railway-systems pages. The per-grade axle-load limits, the turnout speed limits, the minimum lot, the discount and approval thresholds, and all prices are modeled from comparable products and are not published by voestalpine, so those figures are illustrative.

The AI sales agentBuilds the rail and turnout 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
Rail grade against track typeOnly with care
Grooved grades go only with grooved track and the other grades only with Vignol track, so a rule must test the grade family against the track type in both directions.		Grade and track type are fields of one request and are compared on every call.
An axle load that every part must carryOnly with care
The same axle load is limited by the rail grade and, separately, by the turnouts, so two rules must read one figure and a quote can fail both.		Each limit is its own finding and both are returned together.
Turnout speed limits that depend on the turnout typeOnly with workarounds
Standard turnouts, grooved turnouts and high-speed turnouts each have a different diverging-speed limit, so the rule must branch on the type.		The type and the speed sit in one request, so the right limit is applied on every call.
Options that need other items on the quoteOnly with workarounds
Intelligent Turnout monitoring needs turnouts, and a turnout type needs a quantity, so a rule must look across several fields at once.		Quantity, type and option are checked together whatever order they arrive in.
A stated gross that must match the itemsOnly with workarounds
The gross is tonnes times the grade price plus turnouts times their unit price and the monitoring add-on, so a separate rule must recompute it and compare it with the stated figure.		The stated gross is a field of the request and is compared with the computed gross on every call.

  

  

    
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 axle load is 28 t; the quoted rail grade is modeled for at most 25 t.
2Turnouts are modeled for axle loads up to 40 t; the quote has 42 t.
3400 UHC HSH is stated for -40 to +50 degrees C; the site range is -45 to 40.
4Heavy haul needs 350 HT HSH or 400 UHC HSH rail; the quoted grade is not modeled for heavy haul.
5High-speed lines need 350 HT HSH, 340 Dobain HSH or 400 UHC HSH rail; the quoted grade is not modeled for high speed.
6The quoted rail grade is not made for this track type; grooved grades go with grooved track and the other grades with Vignol track.
7The turnout type does not fit the track type; grooved turnouts go with grooved track, standard and high-speed turnouts with Vignol track.
8The diverging speed of 170 km/h is above what this turnout type is modeled for (standard 160, grooved 50 km/h); use a high-speed turnout.
9High-speed turnouts are built with 350 HT HSH, 340 Dobain HSH or 400 UHC HSH rail; the quoted grade is not.
10Rail is quoted in lots of at least 100 tonnes; the quote has 50 tonnes.
11The quote has neither rail nor turnouts.
12The turnout type and the turnout quantity disagree (quantity 0).
13Intelligent Turnout monitoring is ordered but the quote has no turnouts.
14A discount of 10 percent needs sales-manager approval.
15A discount of 20 percent is above the 15 percent limit.
16A custom rail profile or track gauge needs engineering review before the quote is sent.
176,000 tonnes of rail needs rolling-slot confirmation from the mill.
18The quote states a gross of 1,300,000 EUR; the quoted items add up to 1,330,000 EUR.
19The axle load is 45 t; the quoted rail grade is modeled for at most 25 t.
20Turnouts are modeled for axle loads up to 40 t; the quote has 45 t.

  

  

    
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. Heavy haul, axle load and totals: A heavy-haul quote on R260 rail at 45 t axle load with standard turnouts, a 20 percent discount and a wrong gross

    
The agent's request. Please quote 1,000 tonnes of R260 rail for a heavy-haul line on Vignol track, with an axle load of 45 t and site temperatures from -10 to 40 degrees C, plus 4 standard turnouts with a diverging speed of 180 km/h, with a 20 percent discount, for a gross of 950,000 EUR. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Axle load above the grade limit. The axle load is 45 t; the quoted rail grade is modeled for at most 25 t.
    • 
      
  • Heavy haul needs a heavy-haul grade. Heavy haul needs 350 HT HSH or 400 UHC HSH rail; the quoted grade is not modeled for heavy haul.
    • 
      
  • Turnout axle load above 40 t. Turnouts are modeled for axle loads up to 40 t; the quote has 45 t.
    • 
      
  • Diverging speed above the turnout type limit. The diverging speed of 180 km/h is above what this turnout type is modeled for (standard 160, grooved 50 km/h); use a high-speed turnout.
    • 
      
  • Discount above 15 percent. A discount of 20 percent is above the 15 percent limit.
    • 
      
  • Stated gross does not match. The quote states a gross of 950,000 EUR; the quoted items add up to 1,080,000 EUR.
    • 
    

    
2. Heavy haul, axle load and totals: The agent resubmits after fixing the errors

    
The agent's request. Please quote 1,000 tonnes of 400 UHC HSH rail for a heavy-haul line on Vignol track, with an axle load of 40 t and site temperatures from -30 to 45 degrees C, plus 4 standard turnouts with a diverging speed of 120 km/h, with a 5 percent discount, for a gross of 1,680,000 EUR. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Quote summary. Gross 1,680,000 EUR (1,500,000 EUR rail, 180,000 EUR turnouts), discount 5 percent.
    • 
    

    
3. Tram track and turnouts: A tram quote with 340 GHT HSH rail on Vignol track and grooved turnouts at 60 km/h

    
The agent's request. Please quote 500 tonnes of 340 GHT HSH rail for a metro and tram line on Vignol track, with an axle load of 12 t and site temperatures from -10 to 40 degrees C, plus 3 grooved turnouts with a diverging speed of 60 km/h, for a gross of 864,000 EUR. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Rail grade does not fit the track type. The quoted rail grade is not made for this track type; grooved grades go with grooved track and the other grades with Vignol track.
    • 
      
  • Turnout does not fit the track type. The turnout type does not fit the track type; grooved turnouts go with grooved track, standard and high-speed turnouts with Vignol track.
    • 
      
  • Diverging speed above the turnout type limit. The diverging speed of 60 km/h is above what this turnout type is modeled for (standard 160, grooved 50 km/h); use a high-speed turnout.
    • 
    

    
4. Tram track and turnouts: The agent resubmits after fixing the errors

    
The agent's request. Please quote 500 tonnes of 340 GHT HSH rail for a metro and tram line on Grooved track, with an axle load of 12 t and site temperatures from -10 to 40 degrees C, plus 3 grooved turnouts with a diverging speed of 30 km/h, for a gross of 864,000 EUR. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Quote summary. Gross 864,000 EUR (750,000 EUR rail, 114,000 EUR turnouts), discount 0 percent.
    • 
    

    
5. High-speed turnouts and monitoring: A high-speed quote with R260 rail, a high-speed turnout type, no turnouts and Intelligent Turnout monitoring

    
The agent's request. Please quote 400 tonnes of R260 rail for a high-speed line on Vignol track, with an axle load of 20 t and site temperatures from -10 to 40 degrees C, plus 0 high-speed turnouts with a diverging speed of 220 km/h and Intelligent Turnout monitoring, for a gross of 360,000 EUR. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • High speed needs a premium Vignol grade. High-speed lines need 350 HT HSH, 340 Dobain HSH or 400 UHC HSH rail; the quoted grade is not modeled for high speed.
    • 
      
  • Turnout type and quantity disagree. The turnout type and the turnout quantity disagree (quantity 0).
    • 
      
  • High-speed turnout needs a premium rail grade. High-speed turnouts are built with 350 HT HSH, 340 Dobain HSH or 400 UHC HSH rail; the quoted grade is not.
    • 
      
  • Intelligent Turnout without turnouts. Intelligent Turnout monitoring is ordered but the quote has no turnouts.
    • 
    

    
6. High-speed turnouts and monitoring: The agent resubmits after fixing the errors

    
The agent's request. Please quote 400 tonnes of 350 HT HSH rail for a high-speed line on Vignol track, with an axle load of 20 t and site temperatures from -10 to 40 degrees C, plus 2 high-speed turnouts with a diverging speed of 220 km/h and Intelligent Turnout monitoring, for a gross of 844,000 EUR. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Quote summary. Gross 844,000 EUR (460,000 EUR rail, 384,000 EUR turnouts), discount 0 percent.
    • 
    

  

  

    
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 rail operator: rail grade and tonnes, track type, axle load, and more. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the grade, track and turnout rules and returns a verdict, every reason and the totals it computed.
    2. 
      
  2. I will start with the first request by the agent.
    3. 
      
  3. The agent says. Please quote 1,000 tonnes of R260 rail for a heavy-haul line on Vignol track, with an axle load of 45 t and site temperatures from -10 to 40 degrees C. 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 six findings, all in one answer. The axle load is 45 t; the quoted rail grade is modeled for at most 25 t. Heavy haul needs 350 HT HSH or 400 UHC HSH rail; the quoted grade is not modeled for heavy haul. Turnouts are modeled for axle loads up to 40 t; the quote has 45 t. The other three are on screen.
    6. 
      
  6. The validator returned six 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 turnout speed limits that depend on the turnout type, a rules-based CPQ can do this only with workarounds.
    8. 
      
  8. Now the agent resubmits.
    9. 
      
  9. The corrected request. It changes the rail grade to 400 UHC HSH, the axle load to 40 t, the site temperature, minimum to -30 C, the site temperature, maximum to 45 C and the gross stated on the quote to 1,680,000 EUR. Two more changes are on screen.
    10. 
      
  10. It calls the validator again.
    11. 
      
  11. This time the validator clears it.
    12. 
      
  12. All six earlier findings are cleared.
    13. 
      
  13. Six 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.