What does procurement AI for mining actually include?
Four tools for miners and contractors, and one for suppliers. Each one does a job your stores, procurement or estimating team already does by hand, and each one shows its working.
| Tool | Who it is for | What it does |
|---|---|---|
| Stock level (min/max) engine | Miners and contractors | Recalculates min and max for every stock line from net usage, real lead times and criticality, and shows why each level moved. |
| PO vs contract checker | Miners and contractors | Checks every purchase order line against the rate that applied on the day it was raised, and flags anything above it. |
| Contract price book | Miners and contractors | One price list built from your own contract history: what was agreed, what was last paid, and when each rate applied. |
| Commercial assistant in Microsoft Teams | Supply chain and commercial leaders | Answers questions like "what did we spend with this supplier last quarter" from your own purchasing, receipt, payment and stock data. |
| Quote Desk | Mining services suppliers | Reads RFQ emails, drafts the quote from your price book, and checks every PO that comes back against the quote before anything ships. |
None of these replace your ERP. They sit beside it, read from it, and hand your people a short list of things worth acting on.
Are you a miner or a supplier, and where do you start?
Both sides start the same way: a free 15 minute call with Tristan, or a quick enquiry. If there is something worth fixing, you get a clear scope and price for the build. What gets built depends on which side you are on.
| Miners and contractors | Mining services suppliers | |
|---|---|---|
| Who it suits | Mid-tier miners and large mining contractors. Supply chain, procurement, stores and operations managers. | WA suppliers with 10 to 100 staff who quote off RFQs every week. Owners, GMs and estimators. |
| How it starts | Free 15 minute call, then a clear scope and price | Free 15 minute call, then a clear scope and price |
| What gets built | Stock level (min/max) engine, PO vs contract checker, contract price book and an assistant in Microsoft Teams | Quote Desk: reads RFQ emails, drafts quotes from your price book, checks every PO against the quote. Optional stock levels for your own warehouse. |
| How it's priced | Consulting day rate, agreed after a 15 minute call | Consulting day rate, agreed after a 15 minute call |
Read more on spare parts optimisation, PO vs contract rate checking or Quote Desk for mining suppliers.
How does an engagement run, step by step?
A free call first, then a clear scope and price, then the build. Work is charged at a consulting day rate. The price depends on your sites, systems and data, so Tristan talks to you first, then sets out what he'd build, how many days it takes and what it costs, before any work starts.
- 15 minute call. Free, no obligation, no pitch. Tristan asks what you run and what is costing you, and tells you straight whether there is something worth fixing and roughly what it would cost.
- A look at a sample export. Only if it is useful. A read-only sample of your stock, PO, contract or quote data shows whether the data can support the build.
- A clear scope and price. What gets built, for which sites or inboxes, and the price, based on the days the work needs, in writing.
- Build inside your systems. Your Azure, Databricks or Snowflake for miners. Your Microsoft 365 or Google account plus an Australian-hosted database you own for suppliers.
- Run, check and support. Every number links back to the row it came from, so your team can check it. Support covers monthly re-runs, fixes, new sites and new questions.
What results has this produced so far?
These tools were first built inside an international tier 1 mining operator. These are the results that can be shared.
- The stock level engine matched the site inventory expert's own spreadsheet 99.8% of the time.
- One rule change, netting returns off usage, took agreement on disputed items from 73% to 98.4%.
- A commercial assistant in Microsoft Teams answers from millions of purchasing, receipt, payment and stock rows in about a minute. Commercial leadership use it without being asked to.
- Every answer passes 511 automated checks, and every mistake a user catches becomes a permanent test.
- It caught a whole class of wrong spend numbers. Quotes and unapproved orders were being counted as spend. One rule fixed it.
- A tender scorer calibrated against a real evaluation panel matched the panel exactly on one bid and came within a point on another.
See the min/max engine, the Teams commercial assistant and the verification method case studies, or read the full mining supply chain case study.
Why does AI procurement lose trust, and what is different here?
AI procurement tools lose trust fast when the AI is asked to produce the numbers. It guesses, nobody can check it, and the team goes back to their spreadsheets. Here the maths is fixed and the AI only explains it.
- Numbers come from rules, not a language model. Min/max, contract variance and quote totals are calculated the same way every time.
- Every number links back to its line. A flagged PO shows the PO line, the contract rate and the date that rate applied.
- A person approves. The AI never changes a stock level, approves a PO or sends a quote on its own.
- The data gets audited first. Bad data gives confident wrong answers, so the first job is finding what is broken. See the AI data audit case study.
- Savings are described honestly. Cutting excess stock is released working capital over time, not instant profit. Overpaid POs are money to recover or cost avoided from here on.
What do you need to provide?
Nothing for the first call. For the build: exports from your systems, one contact who knows the data, and a few hours of a subject expert's time to check findings.
| Miners and contractors | Mining services suppliers |
|---|---|
| Stock on hand by line and site, with current min and max | Recent RFQs, as emails or PDFs |
| 12 months of issues and returns | The quotes you sent back for each one |
| 12 months of PO lines with approval status | Current price lists and supplier costs |
| Contract rates with effective dates, and lead times if you hold them | Your standard terms and quote template |
CSV or Excel straight out of your ERP is fine. So is read-only access to your data platform.
How does your data stay safe?
Your data stays in your own environment. For miners, the build runs in your Azure, Databricks or Snowflake. For suppliers, it runs in your Microsoft 365 or Google account plus an Australian-hosted database that you own.
Access to source systems is read-only. Nothing writes back to your ERP, and nothing is approved, ordered or sent without a person signing it off. If you stop working with Source Digital, your data and your database stay where they are, in your account.
Every service and guide in this section
Start with the first step that matches your side of the supply chain, or read the guide for the problem in front of you. Each guide teaches the method properly, with the formula, a worked example and the mistakes that matter on a mine site.
Services
- Quote Desk: RFQ to Quote in Minutes. Quote automation for WA mining suppliers. Quote Desk drafts quotes from your price book and checks every PO against the quote. Book a free 15 minute call.
- Spare Parts Optimisation for Mining. Spare parts optimisation for mining: min/max set from net usage, real lead times and criticality. Matched the site expert 99.8% of the time.
- Every PO Checked Against the Contract Rate. Purchase order matching against contract rates, not just three way match. Every PO line checked against the rate that applied on the day.
- A Contract Price Book That Stops Leakage. A contract price book built from your own contract history, so every PO and quote is checked against the agreed rate. Find and stop contract leakage.
By team and region
- AI for Mining Stores and Warehouses. AI for mining stores and warehouses: min/max levels set from real usage, dead and slow stock found fast, and parts shared across sites. A person approves.
- AI for Mining Procurement Teams. AI for mining procurement teams: every PO checked against the contract rate, spend numbers you can trust, and answers in Microsoft Teams in about a minute.
- Mining Supply Chain AI in Western Australia. Mining supply chain AI from Perth for miners, contractors and suppliers across the Pilbara, Goldfields and Mid West. Stock levels, PO checks, fast quotes.
- Inside a Tier 1 Miner's Supply Chain AI. Mining supply chain AI case study: what was built inside an international tier 1 mining operator, how it was proven, and the results. 99.8% agreement.
- AI for mining services suppliers. Our wider page on AI for mining services companies: RFQ automation, parts catalogues, compliance documents and customer portals.
Stock and spares guides
- How to Set Min/Max Levels and Reorder Points. How to calculate reorder point and set min max inventory levels for mining spares, with a worked example, Excel formulas and mistakes to avoid.
- Safety Stock Calculation for Mining Spares. Safety stock calculation for mining spares: the formulas, a worked example, lumpy and critical spares, Excel steps and the mistakes sites make.
- Economic Order Quantity for Mining Spares. Economic order quantity for mining spares: the EOQ formula, a worked example, price breaks, Excel steps and when EOQ is the wrong tool for a spare.
- Inventory Turnover Ratio for Mining Spares. How to calculate inventory turnover ratio for mining stores: the formula, a worked example, Excel steps and the mistakes that skew spares turnover.
- How to Clear Dead Stock and Obsolete Spares. How to find dead stock, obsolete stock and slow moving inventory in a mine store, with a worked example, Excel formulas and what to do with each line.
- Critical Spares and Rotable Spares Explained. What critical spares and rotable spares are, how to decide which to hold, and how to size a rotable pool. Worked examples with the maths shown.
Procurement and supplier guides
- The MRO Procurement Process on a Mine Site. The MRO procurement process for mining, step by step: requisition, sourcing, PO, receipt, three way match and review, with a worked price check example.
- Maverick Spend and Three Way Matching. What maverick spend is, how three way matching works, and why a three way match still lets you pay above contract rate. Worked examples included.
- How to Write a Quote From an RFQ. How to write a quote from an RFQ for mining customers: read it, price each line, state terms and validity, then check the PO. Worked example included.
- Vendor Managed Inventory for Mining Suppliers. Vendor managed inventory and consignment stock for mining suppliers: how to set stock levels, invoice on consumption and stop losing margin on site stock.
