Ensuring Uptime and Reducing Risk on the Copenhagen Gas Network

Case Background

HOFOR is the operator of Copenhagen’s city gas supply network. The grid consists of a low-pressure network to 300,000 private households, that supplies town gas from a pressure ring through several regulator stations. At the time of this case the pressure ring was supplied from two gasworks, with two more being under construction.

For HOFOR, the main challenge is to ensure full availability of the gas supply network, at the lowest risk possible.

The grid is 80 km with approx. 300 mbar and 800 km with 10 mbar. The 10 mbar grid is not able for purging in case of a major encroachment of air; meaning a major risk for HOFOR (and the city of Copenhagen) is pressure drop in the city gas network. This may allow air to enter the pipes, which would lead to:

  • Safety issues, as flammable situations may occur, and

  • Huge economic consequences, in case the gas network must be taken down (estimates costs between 270 and 400 Million euros).

What Was Done

The first step was to define the acceptable risk for catastrophic situations. In this case, that encompassed using the Mean-Time-To-First-Failure (MTTFF) to calculate the return period (likelihood) of catastrophic events. As Copenhagen is close to water, two key inputs here were the dangers of high sea water and heavy rain. Rambøll and HOFOR then established a matrix based on return period, and the economic consequence (cost in MDKK).

Economic Risk Matrix: Current system

The illustration was alarming to HOFOR, and it was clear that the risk picture needed to be improved. The most tangible result was to remain in the same consequence class, but reduce the likelihood significantly. HOFOR had decided to build 2 more production plants, and calculated how much that would affect the result in the same matrix.

Next step was the Flow Based Simulation, using Miriam RAM Studio. Rambøll created the reliability block diagram and applied advanced modelling flows, including input of supply rates, failure rates, and repair times. By doing this, the team got:

  • Visual insight into the supply network

  • Illustration of flows, and effect of failures on re-routing of gas flow

  • Detailed evaluation of breakdown events, combined events, and frequencies

Overview of complete RAM model

Sub-model of gasworks with three production lines

With the RAM model representing the designed system, the team proceeded to the next step of supply network simulation. Miriam RAM Studio uses Monte Carlo simulations to establish expected system availability and reliability, based on multiple replications generating random events during the defined system life time. Rambøll simulated 20,000 replications of 100 years of operation.

Model input & Failure Event Timeline

For each single replication, the results can be explored visually, and time stepped for insight. For Rambøll, this was crucial for quality assurance of the model, and to get a clear understanding of failure and repair events.

On-the-Fly simulation in Miriam RAM Studio

Furthermore, the flexibility of Miriam RAM Studio facilitated an iterative process between Rambøll and HOFOR. This allowed the team to achieve greater learning during the project, and to refine parameters and data as new insight was gained.


Conducting the work with flow simulation in Miriam RAM Studio allowed HOFOR a solid argument for the planned expansion of two more production plants. HOFOR had convinced the network owner to conduct the expansions, and the study results documented the importance both financially and risk.

Economic Risk Matrix: 2 Additional Plants

Economic Risk Matrix: 2 Additional Plants

Other significant results include:

  • Modified construction design for 3 regulator stations to ensure no gas flow backward in case of total failure on electricity. That has resulted in an overall return period of >100,000 years for the most critical situations, proving to HOFOR and the stakeholders that risk has been significantly reduced.

  • Upgraded power supply to the largest plant to have independent 10 kV and 400 V, and calculated cost-benefit effects of back-up generators.

  • Gained a visual understanding of the supply network, and the consequences component reliability and single failures impose to the system.

“We have backed up large financial investments with analysis from Rambøll using Miriam RAM Studio.”

Jørgen S. Jeppesen
Sr. Consultant, City Gas


Next step for HOFOR and Rambøll is the evaluation of going from 4 to 3 under-water supply lines. As one is expected to be shutdown soon, HOFOR wants to evaluate the system effect, and what actions need to be taken to ensure that the low risk-level is maintained.

This study was performed for HOFOR by Rambøll.



Got any questions, or want a
free trial of Miriam RAM Studio?


Fill out this form and we'll get in touch!

Sign up for news and updates on Miriam RAM Studio and/or ExproSoft:

A product by


Miriam RAM Studio is developed and distributed by ExproSoft AS, HQ based in Trondheim, Norway.

Read more at www.exprosoft.com.

  • LinkedIn - White Circle
  • Facebook - White Circle

© 2017 by ExproSoft AS