The industrial safety practice
Industrial safety monitoring in Saudi Arabia
For HSE teams whose observation programme depends on how much a supervisor happens to see: Interkey implements camera-based safety monitoring on Saudi industrial sites, on the CCTV the site already owns, with the video processed in-Kingdom and the alerts landing in the safety workflow that already exists.
The problem
The coverage problem every safety programme has
A serious HSE programme already knows what to look for: the unsafe acts and unsafe conditions its observation cards record, the PPE matrix its inductions teach, the zones its barriers mark. What it cannot buy, at any headcount, is coverage. A supervisor sees a fraction of one area of one shift, behaviour changes when the supervisor is present, and the leading-indicator numbers the programme reports are made of whatever fraction happened to be witnessed.
Meanwhile the site already has hundreds of cameras, recording diligently and watched almost never. Interkey’s industrial safety practice connects those two facts: the existing camera estate becomes a continuous observation layer for the conditions the safety team already defines, PPE compliance, restricted-zone entry, unsafe proximity to moving equipment, with detections routed into the HSE workflow the site already runs and counted into the indicators it already reports.
Three things this is not, to save anyone an unnecessary meeting: it is not a fire and gas system, it is not an ISO certification consultancy, and it is not employee surveillance, what is detected is defined safety conditions, governed by rules the site’s HSE and legal owners set before go-live.
Responsibilities
Who does what: the honest division of labour
Three parties make this work, and deployments fail where their boundaries blur. The detection technology is SquintPRO, Interkey’s own computer-vision product; Interkey is also the implementation and operations party; the safety authority stays exactly where it is.
| Interkey does | The product does | Your site keeps |
|---|---|---|
| The camera survey: which views are usable for detection, which need adjustment, which never will be | Detection itself: PPE, zone entry, proximity events, on the streams the survey qualified | The definition of what matters: which conditions, which zones, which PPE classes, per site rules |
| Integration: alerts into the VMS, control room and incident system the site already uses | The detection record: timestamp, camera, zone, event class, evidence frame | The response: who acts on an alert, and what the escalation is, your procedures, not a vendor default |
| Tuning against real site conditions, and operation after go-live | Natural-language search across footage when an investigation needs it | Safety authority and accountability, which no technology vendor can hold for you |
Before you call anyone
The camera-suitability self-check
Apply this to your own estate before talking to any vendor, including Interkey. It is the fastest way to know whether this class of technology is worth your time, and it is the first thing a survey will establish anyway.
- Do cameras actually cover the areas where your recordable events happen, or only gates and perimeters?
- At the distances that matter, can a human freeze a frame and tell whether a worker is wearing a hard hat? If a person cannot, software cannot
- Are key views usable across the shift, or does midday glare, backlight or night lighting wash them out?
- Are the cameras reachable as network streams, or locked inside a recorder nobody can integrate with?
- Is there somewhere on site, or in your network, where video processing could run, if footage must not leave?
- When an alert fires at 02:00, is there a control room, duty phone or supervisor who would own it?
First step
How a first engagement is scoped
Not with a purchase order. The entry point is a camera-suitability review of one site, which costs you a site walk and some stream access, and answers the only question that matters first: whether your existing estate can support detection at all, and where. If the answer is yes, the next step is a properly scoped pilot, one site, defined cameras, one or two detection cases, success criteria written before it starts, because the only detection numbers worth believing are the ones measured on your own cameras in your own conditions.
Go deeper
The industrial safety library
PPE detection on the cameras you already own
Detectability by PPE class, camera placement, false positives, privacy, and what happens when an alert fires.
Industry guideAI safety monitoring for oil and gas sites
Hazard scenarios, classified areas, permit-controlled work and contractor workforces: what changes on a hydrocarbon site.
Evaluation guideScoping a pilot that proves something
Success criteria, measurement discipline, and the honest failure modes of computer-vision safety pilots.
ProductSquintPRO, deployed by Interkey
The detection product itself, and what Interkey’s deployment work around it involves.
Buyer questions
What buyers ask us
Direct answers to the questions that come up in real evaluations. Anything missing, ask us at the bottom of the page.
Who actually implements and supports this inside Saudi Arabia?
Interkey does, from Riyadh: the survey, the integration with your VMS and incident systems, the in-Kingdom or on-premise processing deployment, the tuning period, and operation afterwards, on the Sunday-to-Thursday week, in Arabic and English. The detection product is SquintPRO, Interkey’s own; the delivery responsibility is also Interkey’s.
Can this run on our existing CCTV, or is a new camera estate required?
It is designed for existing estates, and the survey exists to establish which of your cameras are usable as-is, which need adjusting, and which views genuinely need new hardware, normally a subset, not the estate. A proposal that begins by replacing functioning CCTV is usually a proposal that does not proceed, and deserves not to.
How is worker privacy handled on a monitored site?
As a design input with named owners: retention, footage access, masking and how the workforce is informed are set with your HSE and legal teams before go-live, not defaulted by a vendor, and processing can stay on site entirely. The full design treatment is in the PPE guide’s privacy section; for the legal position itself, bring your counsel in early — that is their call, not a vendor page’s.
How does this handle contractors rather than direct employees?
Detection does not care whose badge a worker carries, which is precisely its value on multi-employer sites, where enforcement runs through contracts rather than employment. What needs designing is the workflow — routing to the contractor’s own supervision and feeding the contractor-performance picture your site already maintains — covered concretely in the oil and gas guide’s contractors section, where multi-employer sites are the norm.
Can detections serve as audit evidence for our ISO 45001 system?
Detection records, timestamped, zoned and classified, are exactly the kind of objective evidence a management-system audit values, and they generate continuously rather than in the week before the audit. How they enter your evidence trail is part of the integration design. What they are not is a certificate: no monitoring technology confers compliance, and a vendor implying otherwise is selling something.
What does deployment cost depend on?
Site count, usable camera count, the detection cases in scope, where processing runs (on-premise hardware versus existing infrastructure), and how much integration your alert workflow needs. Quotes are per site on those variables. The honest first spend is small: a suitability review that tells you whether the rest is worth pricing.
Next step
Start with a camera-suitability review
Describe the site and what the last audit flagged — the practice conversation starts with your cameras and your rules, not a product demo.
Or directly
+966-11-2180999 info@interkey.com.saTawuniya Towers, North Tower, 7th Floor, King Fahad Highway, Olaya, P.O. Box 56835, Riyadh 11564, Saudi Arabia
Elsewhere