Predictika · Aerospace fasteners

An AI sales agent's lockbolt order, checked before it goes out

An AI sales agent prepares an order for titanium shear lockbolt pins with their collars for a composite structure: the pin size and grip dash, the thickness of the stack being joined, the collar and swage gauge, the loads each fastener must carry, and the total shear strength it states. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the specifications Howmet publishes in its online catalog and returns a verdict, every reason and the figures it computed.

Scope: LGP composite titanium shear lockbolt pins (LGPL9SP protruding head, 5/32 to 1/2 inch) with SLFC-MV flanged swage-locking collars: listed part numbers, grip range, collar, swage gauge, shear and tension strength. See everything this validator covers

This validator models one fastener line of the public Howmet Fastening Systems catalog: the LGP Pin Composite, Protruding Head (LGPL9SP, 6AL-4V titanium, 95 KSI ultimate shear) with its SLFC-MV flanged lightweight shear-type swage-locking collar. It checks any order of these fasteners, not just the scenarios shown below.

  • Listed part numbers — eight pin sizes (5, 6, 7, 8, 10, 12, 14 and 16 thirty-seconds of an inch), each listed only for certain grip dashes: dash 2 to 15 for sizes 5, 7 and 8, dash 2, 3 and 5 to 15 for size 6, dash 3 to 14 for size 10, and dash 4 to 14 for sizes 12, 14 and 16. 100 part numbers in all.
  • Grip range — the thickness of the stack being joined must fall inside the design grip range of the chosen grip dash (for example dash 7 covers 0.376 to 0.438 inch).
  • Matching hardware — the catalog names one SLFC-MV collar and one HG118 pin position swage gauge for each pin size, and the quote must list the ones for its own size.
  • Strength — the design shear load per fastener against the pin's minimum double shear strength, the design tension load against the minimum ultimate tensile strength with the listed collar, and the total shear strength the quote states against quantity times the catalog value.

Any order of these pins and collars that touches these rules is in scope — not only the scenarios pictured here. Not covered: prices (the catalog shows none, so this page has no price build-up), other LGP pins (metallic applications, flush or countersunk heads, tension types), other Howmet fastener families, the collar variants with a different lubrication, installation tooling beyond the named swage gauge, and hole preparation, which the catalog item pages do not specify.

The AI sales agentBuilds the fastener order 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
Only the pin sizes and grip dashes the catalog lists (100 part numbers)Only with workarounds
The listed combinations are an irregular table: some sizes start at dash 2, some at 3 or 4, one size skips dash 4. A rules-based CPQ needs a compatibility table or a rule per row, and every catalog change means editing them.		Size and dash arrive in one request and are matched to the listed range on every call.
Stack thickness inside the grip range of the chosen dashOnly with care
The range comes from a table keyed by dash, and the thickness is a free entry, so a rule has to look up the range and compare.		The range for the chosen dash is looked up and compared with the thickness on every call.
Collar and swage gauge that must match the pin sizeOnly with care
Easy if the collar is derived from the pin, but when the lines are entered or imported separately each one has to be checked against the pin.		The pin size and the listed collar and gauge are compared together, in any order.
Loads against the strength of the chosen sizeOnly with care
The strengths differ for each of eight sizes and the loads are separate entries, so every size needs its own limits kept in step with the catalog.		The strengths come with the size and both loads are checked on every call.
Total strength the quote statesOnly with care
The stated total is a free-entry figure, so a rule has to recompute quantity times the catalog strength for the chosen size and compare.		The stated total is a field in the request, compared with the computed one 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 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 quote; approval findings need a person to sign off; everything else is computed.

    
1The stack is 375 thousandths of an inch thick; grip dash 7 covers 376 to 438 thousandths of an inch.
2The stack is 439 thousandths of an inch thick; grip dash 7 covers 376 to 438 thousandths of an inch.
3The 6/32 inch LGPL9SP pin is not listed with grip dash 4; the catalog lists it with dash 2, 3 and 5 to 15.
4The 10/32 inch LGPL9SP pin is not listed with grip dash 15; the catalog lists it from dash 3 to dash 14.
5The 12/32 inch LGPL9SP pin is not listed with grip dash 3; the catalog lists it from dash 4 to dash 14.
6A pin size of 9/32 inch is not an LGPL9SP size; the catalog lists 5, 6, 7, 8, 10, 12, 14 and 16 thirty-seconds of an inch.
7The quote lists the SLFC-MV collar for the 10/32 inch pin; the 8/32 inch pin takes the SLFC-MV collar of its own size.
8The quote lists the HG118 pin position swage gauge for the 6/32 inch pin; the 8/32 inch pin is checked with the HG118 gauge of its own size.
9The design shear load is 9,301 lb per fastener; the pin has a minimum double shear strength of 9,300 lb.
10The design tension load is 3,451 lb per fastener; the pin with its listed collar has a minimum ultimate tensile strength of 3,450 lb.
11The quote states 1,800,000 lb of total minimum double shear strength; the quoted fasteners give 1,860,000 lb.
12The quote has 0 fasteners; it needs at least 1.
13The quote lists the SLFC-MV collar for the 8/32 inch pin; the 6/32 inch pin takes the SLFC-MV collar of its own size.
14The quote lists the HG118 pin position swage gauge for the 10/32 inch pin; the 6/32 inch pin is checked with the HG118 gauge of its own size.
15The quote states 1,000,000 lb of total minimum double shear strength; the quoted fasteners give 1,076,000 lb.
16The 14/32 inch LGPL9SP pin is not listed with grip dash 15; the catalog lists it from dash 4 to dash 14.
17The stack is 0 thousandths of an inch thick; grip dash 7 covers 376 to 438 thousandths of an inch.
18The 16/32 inch LGPL9SP pin is not listed with grip dash 3; the catalog lists it from dash 4 to dash 14.

  

  

    
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. Grip range: Stack thickness outside the grip range of dash 5 on a 480-fastener order

    
The agent's request. Please quote 480 LGPL9SP composite titanium protruding-head lockbolt pins of size 8/32 inch with grip dash 5, with the SLFC-MV collar for 8/32 inch pins and the HG118 swage gauge for 8/32 inch pins, for a composite stack 400 thousandths of an inch thick. Each fastener carries a design shear load of 6,000 lb and a design tension load of 2,000 lb. The quote states 4,800,000 lb of total minimum double shear strength. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Stack thickness outside the grip range. The stack is 400 thousandths of an inch thick; grip dash 5 covers 251 to 312 thousandths of an inch.
    • 
      
  • Stated shear strength does not match. The quote states 4,800,000 lb of total minimum double shear strength; the quoted fasteners give 4,464,000 lb.
    • 
    

    
2. Grip range: The agent resubmits after fixing the errors

    
The agent's request. Please quote 480 LGPL9SP composite titanium protruding-head lockbolt pins of size 8/32 inch with grip dash 7, with the SLFC-MV collar for 8/32 inch pins and the HG118 swage gauge for 8/32 inch pins, for a composite stack 400 thousandths of an inch thick. Each fastener carries a design shear load of 6,000 lb and a design tension load of 2,000 lb. The quote states 4,464,000 lb of total minimum double shear strength. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Order summary. 480 fastener(s), each with a minimum double shear strength of 9,300 lb and a minimum ultimate tensile strength of 3,450 lb, 4,464,000 lb of double shear strength in total.
    • 
    

    
3. Loads, collar and gauge: Loads over the 3/16 inch pin strength, with the wrong gauge

    
The agent's request. Please quote 480 LGPL9SP composite titanium protruding-head lockbolt pins of size 6/32 inch with grip dash 6, with the SLFC-MV collar for 6/32 inch pins and the HG118 swage gauge for 8/32 inch pins, for a composite stack 350 thousandths of an inch thick. Each fastener carries a design shear load of 6,500 lb and a design tension load of 2,200 lb. The quote states 3,120,000 lb of total minimum double shear strength. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Wrong swage gauge for the pin. The quote lists the HG118 pin position swage gauge for the 8/32 inch pin; the 6/32 inch pin is checked with the HG118 gauge of its own size.
    • 
      
  • Shear load over the pin strength. The design shear load is 6,500 lb per fastener; the pin has a minimum double shear strength of 5,380 lb.
    • 
      
  • Tension load over the collared strength. The design tension load is 2,200 lb per fastener; the pin with its listed collar has a minimum ultimate tensile strength of 1,900 lb.
    • 
      
  • Stated shear strength does not match. The quote states 3,120,000 lb of total minimum double shear strength; the quoted fasteners give 2,582,400 lb.
    • 
    

    
4. Loads, collar and gauge: The agent resubmits after fixing the errors

    
The agent's request. Please quote 480 LGPL9SP composite titanium protruding-head lockbolt pins of size 7/32 inch with grip dash 6, with the SLFC-MV collar for 7/32 inch pins and the HG118 swage gauge for 8/32 inch pins, for a composite stack 350 thousandths of an inch thick. Each fastener carries a design shear load of 6,500 lb and a design tension load of 2,200 lb. The quote states 3,456,000 lb of total minimum double shear strength. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Wrong swage gauge for the pin. The quote lists the HG118 pin position swage gauge for the 8/32 inch pin; the 7/32 inch pin is checked with the HG118 gauge of its own size.
    • 
    

    
5. Loads, collar and gauge: The agent resubmits after fixing the errors

    
The agent's request. Please quote 480 LGPL9SP composite titanium protruding-head lockbolt pins of size 7/32 inch with grip dash 6, with the SLFC-MV collar for 7/32 inch pins and the HG118 swage gauge for 7/32 inch pins, for a composite stack 350 thousandths of an inch thick. Each fastener carries a design shear load of 6,500 lb and a design tension load of 2,200 lb. The quote states 3,456,000 lb of total minimum double shear strength. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Order summary. 480 fastener(s), each with a minimum double shear strength of 7,200 lb and a minimum ultimate tensile strength of 2,250 lb, 3,456,000 lb of double shear strength in total.
    • 
    

    
6. Listed part numbers: Grip dash 15 on a 5/16 inch pin, which the catalog does not list

    
The agent's request. Please quote 480 LGPL9SP composite titanium protruding-head lockbolt pins of size 10/32 inch with grip dash 15, with the SLFC-MV collar for 10/32 inch pins and the HG118 swage gauge for 10/32 inch pins, for a composite stack 900 thousandths of an inch thick. Each fastener carries a design shear load of 8,000 lb and a design tension load of 3,000 lb. The quote states 6,720,000 lb of total minimum double shear strength. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Not a listed part number. The 10/32 inch LGPL9SP pin is not listed with grip dash 15; the catalog lists it from dash 3 to dash 14.
    • 
      
  • Stated shear strength does not match. The quote states 6,720,000 lb of total minimum double shear strength; the quoted fasteners give 7,008,000 lb.
    • 
    

    
7. Listed part numbers: The agent resubmits after fixing the errors

    
The agent's request. Please quote 480 LGPL9SP composite titanium protruding-head lockbolt pins of size 8/32 inch with grip dash 15, with the SLFC-MV collar for 8/32 inch pins and the HG118 swage gauge for 8/32 inch pins, for a composite stack 900 thousandths of an inch thick. Each fastener carries a design shear load of 8,000 lb and a design tension load of 3,000 lb. The quote states 4,464,000 lb of total minimum double shear strength. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Order summary. 480 fastener(s), each with a minimum double shear strength of 9,300 lb and a minimum ultimate tensile strength of 3,450 lb, 4,464,000 lb of double shear strength 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 an order for titanium shear lockbolt pins with their collars for a composite structure: the pin size and grip dash, the thickness of the stack being joined, the collar and swage gauge, and more. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the specifications Howmet publishes in its online catalog 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 480 LGPL9SP composite titanium protruding-head lockbolt pins of size 8/32 inch with grip dash 5, with the SLFC-MV collar for 8/32 inch pins and the HG118 swage gauge for 8/32 inch pins. Each fastener carries a design shear load of 6,000 lb and a design tension load of 2,000 lb. The rest of the request is on screen. The quote states 4,800,000 lb of total minimum double shear strength.
    4. 
      
  4. Before it goes out, the agent calls the validator.
    5. 
      
  5. Blocked, with two findings, all in one answer. The stack is 400 thousandths of an inch thick; grip dash 5 covers 251 to 312 thousandths of an inch. The quote states 4,800,000 lb of total minimum double shear strength; the quoted fasteners give 4,464,000 lb.
    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.
    7. 
      
  7. What this shows: for stack thickness inside the grip range of the chosen dash, a rules-based CPQ is error-prone for this.
    8. 
      
  8. Now the agent resubmits.
    9. 
      
  9. The corrected request. It changes the grip dash to 7 and the stated total minimum double shear strength to 4,464,000 lb.
    10. 
      
  10. It calls the validator again.
    11. 
      
  11. This time the validator clears it.
    12. 
      
  12. Both earlier findings are cleared, and the validator's figure matches the quote.
    13. 
      
  13. Two 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.