RESTORE is a program management tool that permits the modeling of a required complex set of steps, including time, or costs, to accomplish the repair of a critical infrastructure service, such as the repair of a downed electric line, a broken natural gas main, or downed telecommunications center. However, the logic behind the invention is not limited to critical infrastructures and can be applied to other applications such as the steps (and time and costs) needed to build a house or building, to ship commodities, to launch or build a space shuttle or an airplane, etc. Another application could be an assessment of a military base in which the impact of an outage would be assessed, including estimated down time, costs and resources required to repair the outage, impact on other infrastructures, and alternate repair scenarios. However, the logic behind the invention is not limited to critical infrastructures and can be applied to other applications where PERT or GANT charts are used.
Restore provides the user the ability to easily build a flow chart that "tracks" the repair activities of an infrastructure failure. The flow chart starts with the service/device failure, and then continues with variety of repairs until the service/device is restored/repaired. Parallel paths to complete an action can be added and probabilities for completion can be assigned to each path.
For example, if a oil pipeline pump fails, this failure starts the flow chart with the possible event scenario:
- Notification of the failure,
- Ttravel to the pump to diagnosis the nature of the failure,
- If the site is remote, weather conditions such as a flooded or snow covered road can be included,
- Ordering of parts
- Delivery of parts may take multiple branches, say with electronics being delivered by mail but a new pump needing rail or heavy truck transport,
- Time to repair at the site,
- If electricity is needed at this remote site then running a power line or using a portable generator can be included, and
- Final restored operation.
As an aside, probabilities can be assigned to parallel branches in the model using standard engineering values or the experience of the users. "Common cause" and interdependency failures can also be identified and modeled. For example, if the oil pipeline fails, this may create additional water or telecommunication infrastructure failures. That is, a lack of oil, may result in a diesel electrical generator stopping because of no fuel, and the lack of electricity, may in turn, cause the water pump to stop and the telecommunications system to shutdown.
The software is easy to use and has several features that make an assessment easier:
- A unique "drag and drop" method of adding new nodes or other more complex nodes previously developed to the flow chart.
- The critical path is noted in the flow chart as a highlighted line.
- Other features can be added such as "cost" and multiple parallel branches to repair, complete an activity, can be added.
- Data can be presented in either a tabular or graphical format.
The software is written in C ++ and can operate on a PC or laptop. The GUI is relatively easy to use and self-explanatory. The code is object oriented and libraries of objects are needed. The current critical infrastructure libraries may not be applicable or transferrable to third-party users because of potentially sensitive data. The software has been in use and validated for certain cases using experience from Nalco on gas line repairs. The code can be expanded to develop models with 500 nodes and 499 branches. A Monte Carlo algorithm is used to create random numbers and provide probability outcomes. Typical running times are very fast. For example, a complicated model, with 378 nodes and 499 branches, was run with 10,000 iterations on a 1.9 Ghz PC, with a Pentium 4 chip, and 21,424 KB of RAM, in less than 20 seconds. The time to set up the model depends upon whether or not relevant libraries exist. If the libraries exist, a new model can be set up in several hours or less. In a practical application of an utility gas line repair, it took several utility staff two weeks to determine various scenarios, as the staff had to work through and identify different scenarios and associated probabilities. The inclusion of weather conditions can have a significant affect on the modeling. For example, the resources needed in digging a trench to find a pipeline break in good weather with dry ground is very different if the weather is bad and the ground is wet (and requires additional wall shoring and pumps to pump water out of the trench).
The RESTORE pricing is per copy, as noted in the license for both universities and commercial