Energex is a subsidiary of Energy Queensland Limited, a Queensland State government-owned corporation in Australia. Energex builds, operates and maintains the electricity distribution network in the growing region of South East Queensland.

In late April 2016 Energex embarked upon a project with ORDITAL to collect information on 9,466 Ring Main Units (RMU's) that were distributed across South East Queensland. Instead of sending personnel into the field to enter data into a keyboard or notepad, Energex sent personnel to describe their assets with photos alone using the ORDITAL solution. This is one of several successful projects that Energex has now completed with ORDITAL.

The RMU project ran for just over 2 months and collected 50,524 classified and cataloged photos that were assembled in a 72GB archive. The photos were captured using the ORDITAL mobile app which allows the photos to be cataloged, classified and geo-located with a single click so that field personnel can quickly get in and capture data and move on to the next item. Field personnel did not have to enter any text during the audit - simply taking classified photos to completely describe the RMU.

The ORDITAL solution allows the auditor to capture classified photographs that accurately describe the RMU in a number of important areas:

  • SITE photographs showing where the RMU is physically installed and how it is identified / tagged (typically 2 photos)
  • RMU photographs showing how it is installed, the ID/purchase order, and the manufacturer's nameplate (typically 3 photos)
  • SITE CONDITION photographs showing any conditions that impact the operation of the RMU such as:
    • OVERGROWN VEGETATION
    • UNEVEN GROUND
    • COMPROMISED ESCAPE PATH
    • PHYSICAL OBSTRUCTION

The RMU is geo-located during the audit process.

By taking classified & cataloged photos of the RMU, Energex have fundamentally changed the concept of field data collection. It is no longer about asking auditors to go out and interpret what they see and enter textual information in either written form, or into a hand held device. By taking photos Energex are collecting very detailed "information" - and doing so a lot more accurately and efficiently than traditional text based audits allow.

The raw "information" in the large photo archive can then be turned into structured, actionable data in a variety of ways. For example, specialist Energex personnel could query the database for RMU's in particular scenarios with overgrown vegetation to decide what sort of remedial works are required.

When it comes to turning large amounts of raw "information" into structured data ORDITAL provides crowd sourcing through customized jobs in large crowd sourcing networks to provide structured data of measurable quality. A good example of this might be to get manufacturer information from the photo archive so that this information can be uploaded into Energex's network management systems.

NAME PLATE photograph of Ring Mount Unit

ATTRIBUTE Extracted from Photo
MANUFACTURER ABB
MODEL SAFELINK
CONFIGURATION CFC
SERIAL NUMBER 1YJN36277402
YEAR OF MANUFACTURE 2009
COUNTRY OF MANUFACTURE NEW ZEALAND

When we have classified photographs we understand what attributes we are likely to be able to extract from those photographs. We submit these photos to crowd sourcing networks of millions of people that are available on demand to answer questions about the photos which are submitted back to us as judgments.

If we only have one judgment, as in the case where we send an auditor to enter text in the field, then we have no measure of the accuracy. It might be right, it might be wrong. With the ORDITAL crowd sourcing solution we introduce a concept of measurable quality where an attribute is only accepted when there are multiple agreeing judgments. For example, if we have 2 out of 3 people key in a serial number of 1YJN36277402 we could say we have 67% confidence in the answer, whereas 3 out of 3 would give us 100% confidence. If we only had one person entering that data then we would have no measure of confidence.

The processing of photos is assembled into large batches and processed typically within 24 hours of the photograph being taken with that data then available for Energex.

By far the most expensive part of any audit activity is getting the person physically in front of the asset. The ORDITAL approach is to see taking the classified photograph as the field data capture activity and not text entry. And because all you are asking the field personnel to do is to take photos with an app there is a lot greater buy in from the field, and good quality photos being generated.

The success of these projects by Energex has validated the ORDITAL approach of describing assets with photos. It allowed rapid data capture, faster assembly of structured data, and ensures that network management systems are synchronized with what is physically installed in the field to generate improved asset management outcomes for Energex and their customers.