Workflow Overview for Generalised Multi Run

The workflow for Generalised Multi Run (GMR) is created within the Generalised Multi Run Configuration and this is also where the analysis is defined. The display of the results (grid or GeoPlan) is obtained from the explorer tree context menu (Viewing GMR results: Grid Report and GeoPlan). The GMR white paper has several examples. The first example “Pipe closure and Flow Reversal” shows the basic operation of GMR.

In the beginning you have a simple run such as a Normal Hydraulic run. This is the Base Run. You will create many variations of this simple run. Each variation is called a Test Case.

There could be thousands of test cases, each of which becomes a separate simulation. These test case simulations execute in parallel, making use of the resources of your system.

Each Test Case is identified by its unique name. The Test Case will have at least one instruction to make it different from the Base Run. In its simplest form, each Test Case will have a single instruction and its name will be the name of an object in the network. For example: each test case could take the name of the pipe, and the instruction could be to close the pipe.

A test case can have any name (not just the name of a pipe) and there may be multiple instructions for each test case. (All instructions with the same Test Case name will combine into one run.) You can modify Network fields and Control fields and Run parameters.

The Generalised Multi Run Configuration allows you to create test cases by using a wizard, by generating a grid in a SQL query, and/or by importing appropriate fields from a CSV.

The GMR will generate time varying results for each Test Case. GMR will run an analysis on the results of each Test Case. Usually, you will only keep the summary analysis. (If there are thousands of Test Cases, you are unlikely to look at the detailed time-varying results, and of course large amounts of disk space would be consumed.)

The Generalised Multi Run Configuration allows you to decide whether to keep time varying results. You will usually discard all results and keep the summary only. Alternatively, you can keep all results, or you can execute a query of your choice that will determine whether to retain time varying results any given test case.

After a Test Case run is completed, an analysis is performed. This is governed by the Generalised Multi Run Configuration. For each Test Case, it is possible to save one or more summary result fields for several objects in the network. It is also possible to execute a SQL query after the completion of a Test Case, giving custom summary results. You may also use the Base Run for as results comparison. Additionally, you may also perform Service Level Analysis for critical link analysis simulation.

The summary results are identified by the Test Case name. Note that if the Object ID is set to "global", then the Test Case will be used as the Object ID. This allows test cases to be compared. In addition, it is possible to generate summary values for multiple objects. For example, each Test Case might involve closing a pipe. The analysis could then measure the minimum pressure at each node in a selection list. The outcome is a pressure value for each nominated object (a named node) within each Test Case (a closed pipe). The outcome is a data cube. You can observe how each pipe affects the nodes in the network, and you can observe how each node is affected by all pipes in the network.

The analysis summary is displayed by using the simulation’s context menu in the tree. This is described in Viewing GMR results: Grid Report and GeoPlan. The data cube containing Test Cases and (possibly) Object IDs can be aggregated into grand summary values.

The summary report can be viewed from the standpoint of Test Cases or the standpoint of Object IDs. The most impactful way of viewing the summary results is via a theme in the GeoPlan. The examples of workflows can be seen in GMR white paper