ACHIEVING A RETURN FROM 3D IN OPERATIONS
I started working for a fantastic 3D CAD company back in 1988, when state of the art was a MicroVAX and large connected graphics terminals. Back then 3D CAD was extremely expensive, so the value proposition only stacked up on the largest engineering projects. At the price point of the day nobody was even talking about using 3D in operations.
Over the last 30+ years there have been dramatic reductions in the cost of 3D CAD systems, but no so the cost of the people with the expertise to operate them. Generating accurate 3D models remains a labour intensive process. The technology to build and visualise complex 3D models has been around for 30+ years - where and how it is used has always been about the economics and value proposition for building and maintaining the 3D models.
Engineering, Procurement & Construction
Construction projects have 3 phases - Engineering, Procurement & Construction (EPC). The engineering department builds the 3D model with enough information to generate specifications and requisitions for Procurement to purchase products that meet that specification, then Construction completes “As-Built” documentation to represent the final constructed facility.
During engineering, tagged equipment such as pumps or other packages are modelled generically to enable the production of specifications and requisitions to hand over to Procurement to decide what will be purchased to meet those equipment specifications. The equipment packages are commonly made up of many maintainable items, such as an electric motor, gearbox and pump in the case of a simple pump package. The cost to go back and retrofit the 3D model with the level of detail to represent the As-Built maintainable equipment is significant and the value proposition for most organisations has historically fallen over at this point.
Over the last 10 or so years we’ve seen the emergence of laser scanning of facilities become become readily available, along with the tools to generate accurate 3D plant models from those point clouds.
The value proposition in engineering for this technology has been compelling, particularly when an existing facility is to undergo a new engineering process such as a revamp or expansion. Laser scanning provides a cost effective mechanism to rapidly build an As-Built 3D model upon which to base plant modifications.
Whilst all of this has been going on in the plant engineering space, over the last 5 years we’ve seen a revolution in commercial application of 3D scanning. It is now possible, and far more importantly economically viable, to generate photo-realistic / millimetre accurate 3D models of consumer facilities such as houses. It is now fast and inexpensive - how fast? 25 min to 38 minutes to scan a typical residential home (2000 ft2 / 185 m2). This dramatic reduction in cost for generating 3D As-Built environments has significant potential for making value propositions for 3D in industrial operations become a reality.
Proof of Concept
Just having a pretty picture you can navigate around was only part of the story, but an important part of the story. At ORDITAL our focus is on rapidly capturing asset information by describing your assets with photographs, then generating information that can be uploaded to your Asset Management system to ensure that your asset management strategies reflect the actual assets that are physically installed. If we could combine rapid capture of 3D models and Asset Information it would truly revolutionise the value proposition for 3D in operations.
To test this theory out, ORDITAL partnered with two Melbourne based companies:
Phoria - a software company that are developing technologies to allow for the effective management of consumer scanning at scale.
Guava Insights - an end-to-end innovation consulting and technology delivery organisation..
The Project - Pumping Station
ORDITAL, Guava Insights & Phoria worked together on a pumping station for an Australian Water company. The pumping station was completely scanned in a couple of hours. ORDITAL then captured 87 assets in the pumping station in 86 minutes with 235 photos to describe those assets. The images were then processed to generate high quality asset information that was then linked into the 3D model. The process of linking in this pilot project was not automated, but again was very quick and completed in an hour or so. Take a look a the end result for yourself in the video below.
Augmented Reality is a reality
Effective asset management assumes that you have an accurate portfolio of assets in your database and that you plan and execute asset management strategies against that asset portfolio, and that everyone in the organisation has required visibility into the asset management strategies.
The reality is that organisations don’t have an accurate portfolio of assets in their databases. The level of asset tagging (ID’s) on their physical assets is highly variable. It is not uncommon to see equipment with no tags (id’s) or multiple tags reflecting multiple tagging efforts over many years. All of this leads to ambiguity as to which physical asset matches which asset in your portfolio. If you can’t accurately match your physical asset to your asset portfolio then your asset management strategies are immediately compromised.
To demonstrate this issue I cite one recent project where the client provided me with an export of their asset database for a small plant room. Their database had 70 assets. I walked around around the plant room with the client’s people and 2/3’s of the assets in their database could not be unambiguously identified. It got worse, because in under 90 minutes we collected 120 maintainable equipment items in that plant room that were missing from the database. Nobody had any hope of matching that old database to the physical assets and thus the client’s asset management strategy was completely compromised and they were delivering no value from operating an asset database.
But even in cases where a facility is well tagged and those tags are reflected in the asset database, you still rely on humans to correctly carry out the work against the right tag in the work order, or to record work against the right asset. All of this relies on the technician matching text in the database and text on the asset.
The 3D models and asset tags that are captured during the process outlined above are ready “out of the box” for Augmented Reality applications. As shown in the video, links can be implemented to identify an asset in the 3D model that you are standing in front of (see image below)- then click through to any related information that you desire. Potentially linking directly through to the record in the asset management system and removing the opportunity for incorrectly identifying the asset as the Augmented Reality environment is highlighting what you are looking at.
The technology for 3D has been around for 30+ years, but the value proposition for making it available in operations has always been lacking. Advances in the consumer space, along with ORDITAL’s ability to rapidly capture assets through photos, are turning this value proposition on its head and enabling a new range of solutions such as augmented reality in operations.
We’ve conducted a successful proof of concept which has excited us and those we’ve shared it with. If any of this sounds like it might be of interest to you, please don’t hesitate to contact me (Stephen @ ORDITAL), Simon @ Guava Insights, or Trent @ Phoria