khsundeen

Comparing Workflows in ArcMap to ArcGIS Pro (Part 2): Tracing the Utility Network

Blog Post created by khsundeen on Sep 11, 2017

In other posts, I explored the workflow changes from ArcMap to ArcGIS Pro working in a utility network to edit your data in a versioned environment. With this next post, I wanted to dive deeper into tracing and how it differs from my workflows using the geometric network traces. In this article, I’m still building on tools and data in Esri’s Network Management Beta 1 release.

Network Tracing Compared

The Geometric Network

In ArcMap, geometric network out-of-the-box simple tracing is available for utilities through Esri’s Utility Network Analyst toolbar. Traditional tracing options in the geometric network use network trace weights, which are set when the geometric network is built. These handle the cost of traversing a given path. You have numerous default options for configuring a simple trace. Options include traversing network features by tracing loops, finding paths or tracing upstream or downstream, finding upstream or downstream accumulation using weights, or finding disconnected features. A utility’s network tracing often required complicated and customized solutions for solving the company-specific tracing.

The Utility Network

For the new utility network, tracing uses the concept of a subnetwork through tiers for different asset flows through a system or sub-system. Subnetworks represent any number of tiers or sub-system of how you decide to segment your utility asset flow. For example, in gas there is gathering, transmission, distribution, which have sub-systems (tiers) of a pressure zone within each system. Then, in each pressure zone (tier), there are different isolation zones. In Electric, you have voltage systems like transmission, primary and secondary with different circuits as a sub-system within each voltage system.

Trace Types

Traces are executed and solved against logically and physically-connected features in a domain network. With six default traces (upstream, downstream, connected, subnetwork, subnetwork controllers, and loops), you now have additional summary and filter options of network attribute filters, functions, terminators, nearest assets, and output filters as outlined below. Taken together, you have significantly more control and granularity beyond what the geometric network traces offered.

1. Valid Consistency: returns traced features with a warning if features are included in a dirty area.

2. Functions: returns summaries or other functions on fields attributes such as ADD, AVERAGE, COUNT,MAX, MIN, SUBTRACT.

3. Network Attribute Filters: returns summaries or filters on features with network attributes such as LIFECYCLESTATUS = InService.

4. Terminators: returns features where terminator categories (of the subnetwork or tier) change categories.

5. Nearest: returns features that are closest to the starting trace point based on a user-specified cost (numeric) attribute, categories (categorical), or a combination of Asset Group x Asset Type fields. 

6. Output Filters: returns filtered results from the original selection set. Filters can be terminator categories or Asset Group x Asset Type combinations.

Getting Started with Tracing in the New Interface

In the following examples, I overview the tracing toolbars in ArcMap and ArcGIS Pro 1.4 with the utility network patch. First, let’s detail a few differences between tracing options available for the geometric network and the utility network in Network Management.

Starting with ArcMap tools, I am using the Utility Network Analyst toolbar with my P_PipeSystem_Net geometric network applied for tracing. 

I ran a simple loop trace in both ArcMap and ArcGIS Pro to show which gas pipes looped back on themselves. Besides the obvious user interface differences, the loop trace selected the same features (as it should!) in ArcMap and ArcGIS Pro. I have some layers not displayed at my current zoom scale in ArcMap, which is why the point features don't appear like they do in ArcGIS Pro. 

Understanding Subnetwork Traces

I wanted to identify which features were connected in the distribution network with my starting point at a distribution regulator station. Recall that subnetworks show system tier designations. To understand which systems (gathering, transmission, distribution in gas) are connected where, it’s important to understand the subnetwork trace option. In the following images, I show the results of running a subnetwork trace on the system tier, including containers, and running a SUM function of the SHAPE_LENGTH. My examples are deliberately simple to illustrate the flexibility and additional customization available. 

The subnetwork trace resulted in selecting all connected features within the distribution system. I executed a function on the subnetwork trace to sum all shape lengths for each line feature. Within this distribution subnetwork tier, there are 453 miles of pipe (2,394,397 feet). Further, this subnetwork trace selected only those town border stations that fed into the distribution network. With how my layers are set, I needed to be zoomed in at larger than a 1:60,000 scale to view the mains. 

Tracing with Containers

Recall that in my trace, I selected the option to Include Containers. Containers can be features in the Assembly feature class (in domain network) or Structure Boundary feature class (in structure network). Neither of these feature classes participate in managing the asset flow of gas however. They simply contain features that participate in the flow.

Refresher on the New Data Model

In case you wanted a refresher on the model structure, here’s a basic graphic illustrating both the structure and AssetGroups assigned to each feature class for gas using the UPDM 2017 GASPIPE/UPDM 2017 model. I’ll be adding electric and water soon! AssetGroups are the new Subtype, except that the fields of AssetGroup and its associated AssetType with assigned domains are network-required fields. See RAMTeCH Software Solution's posts on more details of the new model and Network Management.

Getting back to the containers, these container classes only represent that they contain other equipment that does participate in the asset flow. With this trace option, I can choose to select all containers features that contain equipment that participates in the gas flow.

Continuing to explore the system tier and containers brings us to the transfer points between these systems. One transfer point I’m exploring in the Naperville gas system is a town border station, a type of regulator station that drops pressure from the transmission system pressure to a distribution system pressure. Re-examining this subnetwork trace, note that all town border stations were selected between each system subnetwork (indicated by line colors for gathering, transmission, and distribution systems). Notice that all features within the distribution network were highlighted, including those in the container town border station

Mapping the Container Internals

In the utility network, we now have the option to represent all the internal equipment and associate them with a container. You can easily map all the internals of that town border station in your geometric network. In many instances, the internals might not be mapped; this is a simple representation. You can turn off the contents to just view the container to make the map less complex. On the left in this graphic, the “TBS” icon is a simple container representation (as opposed to a detailed representation on the right). 

Wrapping Up

I summarized some of the new functionality of tracing as well as functionality taken forward from the geometric network and ArcMap. All tracing depends on the domain feature class SubnetLine, which is ArcGIS system-managed. It is a mutipart polyline built from all logically and physically-connected features in your domain network feature classes (line, junction, devices) that control the flow of asset (like gas, water, electricity…etc.). The assemblies and structure boundary feature classes can serve as a container to contain the internal equipment, which, even if not displayed on a map, can still be traced to verify connectivity and run more complex analytics along a trace.

Have another question, comment, or idea? Let me know! Working with RAMTeCH experts all the way through your migration to implementation is the way to go. I'm looking forward to seeing you at the upcoming GeoConX conference in Chicago.

Outcomes