- Accidents
- 0
- Claims or scrap
- < 5%
- Standards
- 3
- Cycles
- Cert. + recert.
two consecutive years
the operation's level
quality, environment and OH&S
external audit passed twice
Context
Set Logística SRL runs out of the same warehouse as E-Techlog — two companies, one shared facility. My focus here was the integrated management system and the ERP that holds it up.
The problem
A certified system and a real operation tend to drift apart. The procedure says one thing, the operator does another, and the binder gets updated the week before the audit. When that happens, certification is an administrative cost that improves nothing.
The challenge wasn’t getting certified. It was certifying something people would use every day.
What I did
The tri-standard system
Implementation and certification of ISO 9001, 14001 and 45001 as one integrated system, plus a full recertification cycle — the real test: sustaining the system once the pressure of the first audit is gone.
Integrating all three shares the documentation and the audit cycle, but forces you to resolve where they overlap. Risk is the clearest case: quality looks at failing the customer, environment looks at impact, safety looks at harm to people. Three readings of the same event — forcing them into a single matrix makes all three useless.
Certification was run as a project: scope, schedule, owners and milestones through to the external audit.
The ERP as the backbone
The decision that held everything together: management system records come out of the ERP, not out of parallel spreadsheets. If the operator already logs the movement to do the job, that same record is the audit evidence.
That’s what prevents the drift. Nobody has to “fill in the records” before an audit, because the records are a by-product of the work.
Non-conformities
Managing the full cycle: detection, root cause analysis, corrective action and effectiveness verification. The hard part isn’t raising the non-conformity — it’s verifying months later that the action worked and it didn’t happen again.
The indicators
A management system isn’t held up by the binder — it’s held up by what gets measured. I built the dashboards that tracked the operation: non-conformities opened and closed, response times, incidents, complaints by customer and by root cause. Queried straight against the ERP database with SQL, published in Power BI and Tableau.
That’s what turned management review into a useful meeting: you walk in with the numbers on the table, not with a story.
And it’s where the thread that runs to today starts — the data work didn’t begin in gas infrastructure, it began here, in 2013.
Outcome
- Zero accidents across two consecutive years. In a warehouse operation — forklifts, working at height, constant movement — that doesn’t come from a poster on the wall. It comes from procedure and practice actually matching.
- Customer complaints fell year over year, on an operation already dispatching with under 5% of orders drawing a claim or ending as scrap. That 5% is where the operation stood; the improvement was measured against it.
- Certification and recertification of all three standards passed.
What I took from it
That a management system either helps or gets in the way depending on one decision: where the records come from. If they come from the work, the system improves the operation. If they come from a separate spreadsheet, it’s bureaucracy with a stamp.