home/references/Semafor

// case study

Semafor

Reporting of outages and disconnections of ZSDIS supply points to OKTE

For the largest electricity distributor in western Slovakia, we designed and delivered an application that fully automates the legally required reporting of planned outages and unplanned failures of supply points to the central market system OKTE (the EDC platform). The solution replaced a manual, error-prone and time-consuming agenda with a reliable machine process featuring full auditability — from the detection of a change in the distributor’s database all the way to confirmed delivery to OKTE.

The brief

A distribution grid operator is legally obliged to report outages and failures affecting electricity supply to the central OKTE platform. This process has several demanding characteristics:

  • Machine integration via the OKTE SOAP web service (Downtime of Supply and Delivery Points) with strict security requirements — WS-Security with digital signing and message encryption using qualified certificates.
  • Huge data volume — a single outage can affect thousands of supply points, each identified by a unique EIC code.
  • High reliability demands — no report may be “lost”; an outage on the OKTE side must not cause a message to go unsent.
  • Auditability — the distributor must be able to prove at any time what was sent to OKTE, when, and with what result.

Handling this agenda manually is practically impossible, and the risk of errors or delays is unacceptable from a regulatory standpoint.

The solution

We created a two-layer “Semafor” platform — a high-performance integration backend and a clear web portal for operators.

Fully automatic detection and batching

The backend scans the distributor’s database every minute and recognizes new, changed and cancelled outages and failures on its own (based on a time window and a scan log). For each change it automatically assembles a report, splits large lists of supply points into batches (max. 999 points per message) and prepares them for sending — without any manual intervention.

Reliable delivery at any cost

Communication with OKTE is designed to be resilient to failures. Each message has its own state (ready, sent, awaiting retry, failed) and, in case of failure, is automatically retried with an exponential backoff (up to 10 attempts with a growing interval, up to 15 minutes). Concurrent processing is handled with database-level locks, so the system handles the load safely. If a message cannot be delivered despite all attempts, the system automatically notifies the responsible staff by e-mail.

Security at the level of a regulated market

The integration uses signing and encryption of SOAP messages (WS-Security) with certificates in secure keystores. Operator access is handled by logging in with a company account (KID) via LDAP / Active Directory and secured with JWT tokens.

Full transparency and audit

Every send is recorded completely — including the exact HTTP request and response, APERAK confirmations from OKTE, times and the result. The operator thus has an evidentiary trail for every report and, if needed, can resend a message manually with a single click.

A clear operator portal

The web application provides an overview of outages and failures, the detail of individual batches and messages, sending logs, filtering, pagination and CSV export — all in Slovak, with a visual distinction between the test and production environments.

Benefits

  • Full automation of the legally required reporting to OKTE — without manual work and without the risk of human error.
  • Reliability thanks to intelligent send retries — reports are not lost even during outages on the counterparty’s side.
  • Auditability and compliance — a complete evidentiary trail of every communication with the market.
  • Security matching the environment of a regulated energy market.
  • Proven in practice — the solution has been in production and continuously maintained since 2023.