Predictika · Foundry filter quotes

An AI sales agent's foundry filter quote, checked before it goes out

An AI sales agent prepares a quote for ceramic foam filters for a foundry: the filter type, the alloy being cast, the pore size, the pouring temperature, the number and area of filters, the metal poured per mold and the molding line. Before the quote goes out, the agent sends it to Predictika, which checks it against a declarative model of what fits what and returns a verdict, every reason and the figures it computed.

Scope: Foseco ceramic foam filters (SEDEX, SUPER SEDEX, SEDEX ULTRA, STELEX PrO, STELEX ZR, STELEX ZR ULTRA, SIVEX FC) and KALPUR direct pouring: alloy, pore size, temperature, capacity and options. See everything this validator covers

This validator models foundry filter quotes, not just the scenarios shown below.

  • Filter and alloy — each filter family is offered only for certain alloys: iron, steel, copper or aluminum castings.
  • Pore size and temperature — each filter comes in certain pore sizes and has a pouring temperature limit.
  • Capacity — the metal poured per mold must stay within what the quoted filter area can handle, and the total area stated on the quote must match the filters quoted.
  • KALPUR direct pouring — a direct pouring unit needs a compatible filter for the alloy and a round filter.
  • Approvals — plain SEDEX on a DISAMATIC line, automated placement without a closed-edge filter, and STELEX PrO on a small ductile iron casting need a person to confirm them.

Any foundry filter quote that touches these rules is in scope — not only the scenarios pictured here. Not covered: steel flow control products, feeder sleeves other than KALPUR and other Foseco lines. Product names, filter families and some pore sizes follow the public Foseco website. Temperature limits (except the SEDEX limit), the capacity of each filter, the alloy pairings and the approval thresholds are modeled from comparable foundry filter practice, not published by Foseco. No prices are modeled.

The AI sales agentBuilds the filter quote from the conversation with the foundry 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
Filters that depend on the alloyOnly with workarounds
A compatibility table or one rule per filter and alloy, edited again whenever a filter is added.
The filter and the alloy are checked together on every call.
A temperature limit that depends on the filterOnly with care
Each filter has its own limit, so a rule must look up the limit of the selected filter and compare it with the pouring temperature.
The limit is looked up inside the model and compared on every call.
Capacity from the quoted areaOnly with workarounds
The metal one quote can handle depends on the number of filters, their area and the filter type, so a separate calculation is needed.
The capacity is computed inside the model and compared with the metal per mold on every call.
A stated area reconciled with the quoted filtersOnly with workarounds
The total area comes from the number of filters and their area, and a separate calculation is needed to compare it with the figure the quote states.
The area is computed inside the model and compared with the stated 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.

1SUPER SEDEX or SEDEX ULTRA is the usual choice for vertically parted DISAMATIC molding; a person should confirm plain SEDEX.
2Automated filter placement works best with closed-edge filters (the ULTRA types); a person should confirm the selected filter.
3The selected filter is not offered for this alloy in this model; iron, steel, copper and aluminum castings each use their own filter family.
4The pouring temperature of 1,580 C is above the 1,500 C limit of the selected filter.
540 ppi is not an available pore size for the selected filter.
6121 kg of metal per mold passes through the filters; the quoted filter area can handle 120 kg.
7The quote states 150 cm2 of total filter area; the quoted filters give 100 cm2.
8A KALPUR direct pouring unit holds a round filter inside the feeder sleeve; the quote has a square filter.
9KALPUR direct pouring uses a SEDEX or STELEX PrO filter for iron and steel (SIVEX FC for aluminum, SEDEX for copper); the selected filter and alloy do not fit.
10STELEX PrO is intended for white iron and large ductile iron castings; a person should confirm it for 60 kg of ductile iron per mold.
11At least one filter per mold is needed; the quote has 0.

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. Steel casting with the wrong filter family: Carbon steel poured at 1580 degrees with a SEDEX filter and KALPUR direct pouring

The agent's request. Please quote 1 SEDEX filter, 40 ppi, 60 square centimeters each, for carbon steel poured at 1580 degrees Celsius, with 60 kilograms of metal per mold, using KALPUR direct pouring with a square filter. The quote states a total filter area of 120 square centimeters. It is ready to send.

Verdict: Blocked.

  • Filter not suited to the alloy. The selected filter is not offered for this alloy in this model; iron, steel, copper and aluminum castings each use their own filter family.
  • Pouring temperature over the filter limit. The pouring temperature of 1,580 C is above the 1,500 C limit of the selected filter.
  • Stated filter area does not match. The quote states 120 cm2 of total filter area; the quoted filters give 60 cm2.
  • Direct pouring needs a KALPUR-compatible filter. KALPUR direct pouring uses a SEDEX or STELEX PrO filter for iron and steel (SIVEX FC for aluminum, SEDEX for copper); the selected filter and alloy do not fit.
  • Direct pouring needs a round filter. A KALPUR direct pouring unit holds a round filter inside the feeder sleeve; the quote has a square filter.

2. Steel casting with the wrong filter family: The agent resubmits after fixing the errors

The agent's request. Please quote 1 STELEX ZR filter, 20 ppi, 60 square centimeters each, for carbon steel poured at 1580 degrees Celsius, with 60 kilograms of metal per mold, with a round filter. The quote states a total filter area of 60 square centimeters. It is ready to send.

Verdict: Allowed.

  • Filter summary. 1 filter(s) with 60 cm2 in total, handling up to 120 kg of metal per mold, limit 1,750 C.

3. Too much metal for the filter area: Ductile iron on a DISAMATIC line with 150 kilograms of metal per mold

The agent's request. Please quote 2 SEDEX filters, 30 ppi, 50 square centimeters each, for ductile iron poured at 1420 degrees Celsius, with 150 kilograms of metal per mold, on a DISAMATIC line, with automated filter placement. The quote states a total filter area of 100 square centimeters. It is ready to send.

Verdict: Blocked.

  • Metal per mould exceeds filter capacity. 150 kg of metal per mold passes through the filters; the quoted filter area can handle 120 kg.
  • Plain SEDEX on a DISAMATIC line. SUPER SEDEX or SEDEX ULTRA is the usual choice for vertically parted DISAMATIC molding; a person should confirm plain SEDEX.
  • Automated placement without closed edges. Automated filter placement works best with closed-edge filters (the ULTRA types); a person should confirm the selected filter.

4. Too much metal for the filter area: The agent resubmits after fixing the errors

The agent's request. Please quote 2 SEDEX ULTRA filters, 30 ppi, 60 square centimeters each, for ductile iron poured at 1420 degrees Celsius, with 150 kilograms of metal per mold, on a DISAMATIC line, with automated filter placement. The quote states a total filter area of 120 square centimeters. It is ready to send.

Verdict: Allowed.

  • Filter summary. 2 filter(s) with 120 cm2 in total, handling up to 168 kg of metal per mold, limit 1,500 C.

5. STELEX PrO on a small ductile iron casting: STELEX PrO for a 60 kilogram ductile iron casting

The agent's request. Please quote 1 STELEX PrO filter, 10 ppi, 80 square centimeters each, for ductile iron poured at 1450 degrees Celsius, with 60 kilograms of metal per mold. The quote states a total filter area of 80 square centimeters. It is ready to send.

Verdict: Needs approval.

  • PrO on small ductile iron casting. STELEX PrO is intended for white iron and large ductile iron castings; a person should confirm it for 60 kg of ductile iron per mold.
  • Filter summary. 1 filter(s) with 80 cm2 in total, handling up to 200 kg of metal per mold, limit 1,650 C.

6. STELEX PrO on a small ductile iron casting: The agent resubmits after fixing the errors

The agent's request. Please quote 2 STELEX PrO filters, 10 ppi, 80 square centimeters each, for white iron poured at 1450 degrees Celsius, with 300 kilograms of metal per mold. The quote states a total filter area of 160 square centimeters. It is ready to send.

Verdict: Allowed.

  • Filter summary. 2 filter(s) with 160 cm2 in total, handling up to 400 kg of metal per mold, limit 1,650 C.

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 ceramic foam filters for a foundry: the filter type, the alloy being cast, the pore size, and more. Before the quote goes out, the agent sends it to Predictika, which checks it against a declarative model of what fits what 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 quote 1 SEDEX filter, 40 ppi, 60 square centimeters each, for carbon steel poured at 1580 degrees Celsius, with 60 kilograms of metal per mold. The quote states a total filter area of 120 square centimeters. The rest of the request is on screen.
  4. Before it goes out, the agent calls the validator.
  5. Blocked, with five findings, all in one answer. The selected filter is not offered for this alloy in this model; iron, steel, copper and aluminum castings each use their own filter family. The pouring temperature of 1,580 C is above the 1,500 C limit of the selected filter. The quote states 120 cm2 of total filter area; the quoted filters give 60 cm2. 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 quote are checked against each other.
  7. What this shows: for a temperature limit that depends on the filter, a rules-based CPQ is error-prone for this.
  8. Now the agent resubmits.
  9. The corrected request. It changes the filter to STELEX ZR, the pore size to 20 ppi, the filter shape to round and the total area stated on the quote to 60. It removes the KALPUR direct pouring.
  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 quote. That is the value of Predictika's validator: catching mistakes before they reach the customer. Now you can try the other scenarios yourself.