This document produced by the Open Asset Digital

Twin Working Group (OADT) provides

comprehensive guidelines for the digital capture and
documentation of the physical world to create
accurate and detailed 3D models or representations.
It addresses the challenges posed by multiple
technologies, formats, and data structures, which

REALITY CAPTURE hinder progress towards a Digital Reality ecosystem.

The guidelines aim to assist asset owners and
GUIDELINES Digital Reality practitioners in acquiring high quality
data suitable for the development of Digital Twins.

OPEN ASSET DIGITAL The document includes contributions from major

industry players such as bp, Chevron,
TWIN WORKING GROUP ConocoPhillips, Shell, and ExxonMobil, with
additional input from Petrobras and Saudi Aramco.

April 2025 Content provided by bp, Chevron, ConocoPhillips,

Shell and ExxonMobil with additional input from
Petrobras and Saudi Aramco

Edited by Peter Reynolds

Industry Analyst, and Mark Luciw Co-Chairs of the
Open Asset Digital Twin Working Group and the
OADT Reality Capture Sub-committee.
 OADT Reality Capture Guidelines • April 2025

VISION, EXPERIENCE, ANSWERS FOR INDUSTRY

Open Asset Digital Twin Working Group • 2

 OADT Reality Capture Guidelines • April 2025

Revision Log

Revision Date Editor Description of Changes Status

1 January 27, 2025 P. Reynolds Initial draft Draft

2 Feb 9, 2025 P. Reynolds Separate core from non-core technolo- Draft

gies; added deliverables and neutral file
formats. added accuracy factors, Added
Document contributors.

3 April 24, 2025 P. Reynolds Working Group Review for Publication Publish
version 1

Open Asset Digital Twin Working Group • 3

 OADT Reality Capture Guidelines • April 2025

Table of Contents
Introduction ..................................................................................................................... 5
Document Contributors ............................................................................................................ 6
Distribution .............................................................................................................................. 7
Liability .................................................................................................................................... 7
Scope ....................................................................................................................................... 7
Laser Scanning (LIDAR) .................................................................................................... 8
Fundamentals Principles........................................................................................................... 8
Capture Workflow .................................................................................................................... 8
Post-Capture Workflow ............................................................................................................ 9
Capture Guidance ..................................................................................................................... 9
Deliverables ........................................................................................................................... 14
Accuracy................................................................................................................................. 16
Scan Density and Coverage ..................................................................................................... 19
Target Group and Responsibility............................................................................................. 21
Typical Project Team Members .............................................................................................. 21
Photogrammetry ........................................................................................................... 22
Fundamental Principles .......................................................................................................... 22
Capture Workflow .................................................................................................................. 22
Capture Guidance ................................................................................................................... 22
Deliverables ........................................................................................................................... 24
Accuracy................................................................................................................................. 25
Definitions ..................................................................................................................... 26

Open Asset Digital Twin Working Group • 4

 OADT Reality Capture Guidelines• April 2025

Introduction

This set of guidelines shall be read in the context that the deliverables specified are intended to
reliably support remote ‘as if I were there’ quality data without the need for physical presence.
It has been prepared by the ARC Advisory Group Open Asset Digital Twin Working Group
(OADT) and is designed to provide guidance on the acquisition and processing of 3D reality
capture focused on the development of high-resolution Digital Twins in an open and interoper-
able digital ecosystem. This document should be considered as a guide for specialized service
companies working in the reality capture industry.

The delivered scan and imagery data is intended to consistently meet anticipated use-cases,
which is primarily the dimensional verification of both objects and their spacing. Additional ac-
tivities include completion of fieldwork, conceptual and detailed design, as-found verification,
interference detection, work planning, asset verification, visual context, development of GIS con-
tent, and the creation and updating of high-fidelity as-built 3D models and digital twins.

Digital twins can deliver significant value through the entire asset lifecycle, from design and con-
struction to operations and maintenance, while optimizing and automating associated business
processes. To date, the focus of digital twins has been on the engineering and construction (E&C)
portion of the lifecycle, with technology solution providers making digital twins for engineering
design and building infrastructure. The complex operate and maintain portion of the lifecycle
has been effectively disconnected from design and build stages, leaving huge opportunities for
cost savings in an underserved market.

While digital twins improve collaboration, they need a significant overhaul to become truly
“open.” Openness ensures the interoperability of data and technologies to provide cost-effective,
scalable, and sustainable solutions. Sustainable Asset Digital Twins will give a framework for
connecting dataflows across silos to produce a holistic view of an asset’s information across its
entire lifecycle. Asset owners want to eliminate barriers to integrating and contextualizing dis-
parate data sets while optimizing and automating business processes. Data and software must
be decoupled for value creation. Moving forward it’s the interaction with the data that will be a
key component. What needs to be established are pillars to separate image capture, storing it,
and enriching it with data that can be used anytime, anywhere. Finally, it’s all about acceleration,
adoption in value capture.

About the Open Asset Digital Twin Working Group (OADT)

The Open Asset Digital Twins working group (WG) provides a forum for asset owners whose
purpose is to collaborate, articulate users' needs, and speak in a focused voice to help vendors

Open Asset Digital Twin Working Group • 5

 OADT Reality Capture Guidelines • April 2025

overcome the limitations of digital twins. Together, the WG will align on key industry needs and
challenges and share consistent messaging with vendors on that topic so that they can support
users more effectively. The WG will help vendors see opportunities to play and be profitable in
an "open" approach to the digital twin ecosystem and address gaps in the data and workflows
across the asset lifecycle, which causes losses in productivity, quality, and sustainability.

The OADT are working to:

• Influence suppliers to see opportunities to play and be profitable in an open digital twin
ecosystem
• Amplify a multi-user common voice to suppliers, improve supplier offerings
• Provide continuity in discussions and progress to common goals
• Be an agnostic, 3rd party voice to suppliers, avoiding the legal appearance of any collusion
• Exchange of knowledge (process, discrete, and hybrid industries)
• Access to the market and influence as a group
• Compile best practices, establish critical input/feedback criteria from other users

For more information about the OADT please visit ARC Advisory Group
https://www.arcweb.com/open-asset-digital-twins-working-group

Document Contributors
ARC Advisory Group acknowledges the following content contributors from the participating
asset owner companies.

• Justin Piwetz, BP
• Tony Hunter, BP
• Nick Purday, ConocoPhillips
• Jake Gleason, Chevron
• Seth Taylor, Chevron
• Kyle Daughtry, ExxonMobil
• Michael Hotaling, ExxonMobil
• Nick Pederson, ExxonMobil
• Robert Whittaker, Shell
• Viswa Pinninti, Shell
• Guhan Prahbu, Aramco Services
• Peter Janach, BASF
• Luciano Pereira dos Reis, Petrobras

Open Asset Digital Twin Working Group • 6

 OADT Reality Capture Guidelines• April 2025

Distribution

This paper is offered was created for free and open distribution. The intent of the ARC Advisory
Group Open Asset Digital Twin Working Group is to make knowledge and research readily ac-
cessible to the widest possible audience without any financial or restrictive barriers. This
approach prioritizes the dissemination of information, fostering collaboration, and accelerating
the advancement of understanding within a particular field. By leveraging open access, institu-
tional repositories, and creative commons licenses, authors and institutions empower global
readership, encourage citation and reuse, and ultimately maximize the impact and reach of their
work, contributing to a more equitable and collaborative landscape.

Liability
This paper is provided "as is" without any express or implied warranties, including but not lim-
ited to warranties of merchantability, fitness for a particular purpose, and non-infringement. The
authors and distributors shall not be liable for any direct, indirect, incidental, special, exemplary,
or consequential damages arising in any way out of the use of this paper. Use of the information
contained herein is at the reader's sole risk.

Scope
This specification details the methods, deliverables, quality controls and accuracy to be followed
by a Service Provider for the collection and delivery of reality capture scans to an Operator. The
document can also be used as a reference for internal technology groups within Oil Companies
and EPCs. Technologies included in this document are Photogrammetry, Laser Scanning (LI-
DAR).

Open Asset Digital Twin Working Group • 7

 OADT Reality Capture Guidelines • April 2025

Laser Scanning (LIDAR)

Fundamentals Principles
This specification largely presumes utilization of tripod mounted LiDAR laser scanners generat-
ing structured (Point of View) scan data. However, depending on the asset and use case, other
reality capture technologies may be appropriate to the extent the resulting deliverables meet this
specification. Mobile, SLAM based, LiDAR sensors are rapidly increasing in performance and
represent a viable alternative to static LiDAR acquisition when compromises in data quality can
be made for speed and cost.

Capture Workflow
• Initial Plan
- Identify floor/space plans and the names to be used for the area field in the scans.
- Reference coordinate system monuments to use or be created from Resident Survey Con-
tractor.
- Safety plan including guidelines to ensure no process or safety systems are tripped dur-
ing capture e.g. that camera lighting or laser scanner beam will not inadvertently trip
flame detectors.
- Review previous scans and imagery, if available.
- Wherever possible, walk the area(s) to validate areas of change and scope boundaries.
• Logistics (plan, schedule, permitting, travel).
- Coordinate with Survey Contractor to establish survey control or validate existence and
adequacy for scan registration.
- Mount scan targets as necessary to ensure appropriate coverage throughout the scope
area.
- Targets must be installed before laser scanning but may be surveyed at any time.
- Each scan should capture a minimum of three scan targets within 10 meters that are 135
degrees apart from the viewpoint of the scanner and preferably not in the same horizon-
tal plane.
- All exterior scan targets shall be Surveyed Scan Targets.
- A minimum of three Scan Targets within an enclosed area shall be Surveyed Scan” Tar-
gets.
• Scan and Imagery Capture
- NOTE: For each scan, record the Area name being scanned, Date of scanning, Sequence
number for this area, model of the capture device, serial number of the capture device,
and the device settings as selected by the operator.
- Review the scans for defects before site demobilization.
- Deliver notes, scans, and imagery for post-capture processing.

Open Asset Digital Twin Working Group • 8

 OADT Reality Capture Guidelines• April 2025

Post-Capture Workflow
• Review/cleanup data e.g. remove reflections and out-of-spec data.
- Register (align) capture data to local survey control coordinate system.
- Some scanners support some degree of field registration.
- Registration Validation QA/QC
• Data delivery to Operator
- Preferably, the data should be uploaded directly to Operator’s Cloud Storage System of
Record
- Failing that delivery via USB drive is an alternative
• Reports detailing at a minimum:
- Summary of the business driver for each capture e.g. Work Order # and title.
- Effort – approx. team size and hours expended categorized by field or office activity to
help guide future LOE expectations.
- Issues experienced, any learnings, recommendations for process improvement.
- Spreadsheet containing a record for each scan containing:
- Scan file name.
- Named area of scan
- Model and Serial # of scanner used.
- Scanner settings
- Site status (nominal operation, shut down, excessive temps etc.)
- General Notes on unusual conditions for this scan e.g. heavy vibration, high winds
- Spreadsheet containing a record of:
- Cloud-to-Cloud Registration accuracy report
- Scans to Local Control accuracy report
- Survey Scan Target accuracy report.

Capture Guidance

Scan Intent

• A work order may designate a mix of update and new areas designated for capture. For
each area of scope, it is important to confirm if the request is to capture changes or is for
new bulk capture. Change capture generally requires physical review before scanning to
establish the actual area of change. If available, the use of 3D design models and previous
reality capture imagery of the site can greatly help identify boundaries of scope change.

New Capture / Bulk Capture

Open Asset Digital Twin Working Group • 9

 OADT Reality Capture Guidelines • April 2025

• This process requires the establishment of survey control along with survey control tar-
gets suitable for capture by laser scans and then scanning the entirety of the area in a
manner that ensures achievement of required accuracy, density, and coverage. Unlike
update scanning, the resulting new scans will typically supersede any historical scans of
the area.
• Bulk capture is appropriate for new assets e.g. a new module addition, or when refresh-
ing an existing scanned area due to extensive updates e.g. Fire & Gas sensor retrofit with
new conduit throughout a module may be more cost effectively and reliably captured
through bulk capture rather than several update scans of localized scope.

Update Capture

The process of rescanning to capture visible changes since the last scan capture. To ascertain
the boundaries of scope change, the vendor should review available previous scans, imagery,
and IFC engineering documents of the scope area and use these to help validate actual
changes onsite.

• If there is any doubt about where the changes end, expand the scope area to include sus-
pected areas of change.
• Due to the remote nature of some facilities, the incremental cost for additional scans to
capture suspect areas is trivial compared to the cost and delay for an additional trip to
capture missed changes.

Update capture is not limited to process equipment. Any visible change to the site even if
noted as an “in-kind” replacement should be captured e.g. Valve replacement, replaced
guardrail, relocation of light fixtures.

Be mindful to capture so called “in-kind” and non-process changes that are often not re-
garded as material changes and may not even exist within the site Management of Change
process. Capturing these changes is just as important as process changes, perhaps even more
important as historically they can be the source of field surprise that result in unplanned,
expensive, and delaying field modifications that increase risk.

All updated captured artifacts of a physically changed area must accurately align in the de-
fined coordinate system to properly meld with previously captured scans. Using
permanently installed Survey Control Targets cross-checked with Cloud-to-Cloud registra-
tion with previous scans can be an efficient way to verify the registration accuracy of the
update scans. (See “Scan Registration”)

Scheduling

Open Asset Digital Twin Working Group • 10

 OADT Reality Capture Guidelines• April 2025

• Facility interiors may be done day or night year-round.

• Facility fin/fan or heating/cooling areas should be scheduled at a time of year when the
ambient temperature would be comfortable for people and equipment.
• Exteriors should be done during the day in the summer. Wherever possible interior and
exteriors should be captured at the same time, merging of different surveys can be chal-
lenging as objects may shift between surveys, think Alaska tundra.
• For exteriors, avoid periods when standing water, rain, snow, or fog are present or ex-
pected. These conditions can cause scan occlusions or reflections rendering scans
unusable.
• Generally, capture should occur while assets are in a nominal production status Offline
assets will often experience significant movement from temperature and weight loading
versus their production position.

Walkthrough Scans

• In addition to asset captures, ensure there are eye-level captures within every doorway
and open corridor intersection. These capture points are very useful for orientation, train-
ing, and pathfinding. For stairs and ladders, capture at least the immediate area of the
base and at least one setup on any connecting landing or platform. See note on platforms
under Avoid Blind Spots.

Avoid Blind Spots

• Avoid positioning setups in a linear fashion.

• Vary capture height for asset capture above and below the level of nearby horizontal
or vertical runs of piping or handrails can result in a run of scan data shadowing,
creating blind spots for assets located within the shadow.
• Be mindful of overhead stacked assets e.g. pipe racks, take multiple setups to max-
imize view through gaps. Getting sliver views of each pipe body is helpful, but the
real prize is capturing all connections and changes in direction.
• Be aggressive in capturing views of penetrations through walls, ceilings, and floors.
Frequently there will be structural members nearby and unless the scanner is posi-
tioned carefully, penetration piping and their connecting flanges will not be captured
due to shadowing by these structural members.
• Make sure to capture areas blocked from view by larger assets positioned near walls.
• Higher platforms and walkways are generally high-value positions for capturing the
top of asset details that will otherwise be missed. Often these positions offer only a
view of the top of taller assets e.g. large equipment and vessels, and their connection
points. Multiple setups, even if only slightly repositioned, are worthwhile to capture
these critical areas. Consider scanning at a higher resolution to pick up details of asset
top sides that are at a distance.

Open Asset Digital Twin Working Group • 11

 OADT Reality Capture Guidelines • April 2025

• Be sure to capture piping and other assets under larger assets such as under a larger
vessel, skid, or larger equipment like a compressor. Given the limited vertical view,
be mindful of shadowing by foreground assets in your setups.
• Take note of movable objects e.g. rolling bin, swinging door blocked open, and con-
sider gaining approval from the Facility Operator to temporarily reposition them to
unblock the view of what is behind them.

Registration

The registration process is fundamental to ensuring the dimensional accuracy and usefulness
of the scan data.

• All exterior scans should be registered using Survey Control Targets. If not feasible, an
exterior scan can be registered using cloud-to-cloud to other scan(s) that have been reg-
istered using Survey Control Targets.
• Interior scans can be registered using Cloud-to-Cloud within the local control network,
if the greater of two or 10 percent of the scans are registered to Survey Control Targets
and are physically distributed throughout the interior. For example, in a grouping of 20
scans of an interior area, at least two of the scans must be registered to Survey Control
Targets and these two scans must be reasonably at opposite ends of the interior space.

All transforms introduce dimensional error resulting in varying degrees of uncertainty. With
appropriate survey control this additional error can be minimized, and importantly, quanti-
fied. A tightly aligned survey control network provides both error checking and a quality
foundation to correctly align scan data and understand the limitations of its accuracy. With-
out error checking processes the actual error is hidden and can manifest with unanticipated
consequences of both risk and expense.

Quality

Of the many different aspects of generating and delivering laser scan data there are three that
are fundamental to the resulting quality - and by extension, usefulness - of each scan:

• Positional Accuracy – accuracy of a scanned point measurement relative to its true posi-
tion
• Is largely controlled by the rigor of the registration process, and somewhat by the funda-
mental accuracy and precision of the scanning device and the operator settings.
• It is the responsibility of the capture vendor to choose appropriate equipment, work-
flows, survey controls, and QA process to ensure that all scan data meets the required
positional accuracy.

Open Asset Digital Twin Working Group • 12

 OADT Reality Capture Guidelines• April 2025

• Spatial Resolution/Density – the spacing between adjacent scanned points on a surface

that meets the positional accuracy requirements.
• Controlled by device settings / capabilities for each single sweep at a scan position.
• If the required density can be achieved in a single scan, this typically results in the best
and most consistent quality.
• For small bore e.g. 0.5” assets, scanning from ≤ 3 meters is encouraged to ensure not only
sufficient scan density, but also sufficient imagery detail. Multiple scan positions may
paint the same area, increasing the effective density of scanned points. However, com-
bining multiple scans will inherently increase the error bound (uncertainty) of
measurements and if all scans are positioned far from the target area, none may provide
the desired quality of imagery.
• Coverage - how much of the surface of an asset has been scanned at the minimum re-
quired density of points, each of which meeting the minimum required accuracy.

Each of the fundamental aspects listed above works together to generate a product that is
both dependable and useful.

When making tradeoffs which are the inevitable result of field measurements, positional ac-
curacy relative to the scanned physical surface is primary and cannot be negotiated away.
Between resolution and coverage, coverage will generally have priority as it has a direct bear-
ing on capturing connections and changes in surface direction which involves many use
cases.

Field Equipment

Provisioning of all field equipment used for reality capture is the responsibility of the Service
Provider. Typical items include:

• Capture and Survey Control equipment.

• All equipment to be calibrated per manufacturers recommendations.
• Note model and serial number, and if applicable, firmware version for each tool.
• Power Connection Adapters, Power cables / extensions and/or batteries
• Specialized adapters to intrinsically safe power outlets can be provided by Operator, but
availability must be requested and acknowledged prior to any scheduled scanning.
• Appropriate PPE
• Appropriate semi-permanent attachments e.g. ring to indicate placement of scanning reg-
istration sphere, stencil to prior-approved paint survey control pattern on structural,
paint, paper registration patterns.
• Appropriate temporary Signage indicating to others:
• Scanning is in progress.

Open Asset Digital Twin Working Group • 13

 OADT Reality Capture Guidelines • April 2025

• Advisements e.g. Scanning in Progress - please minimize walking through area.

• Cautions, if any
• Computing device e.g. laptop or tablet
• Software toolchain as needed for capture equipment, data validation, and desired output.

Provisioning of any ladders, scaffolding, or other at-height temporary platforms must be pro-
vided by Operator. Requests for same may require significant lead-time and are not
guaranteed to be honored.

Deliverables

Transport: Preferably upload to owners Cloud System of record. Failing that then provided on
Windows compatible external disk drives with USB 3.0 or newer interface. Disk drives will not
be returned and become the property of Operator.

Geolocation data shall be provided by survey grade GPS equipment. The coordinates shall be
provided on high precision with at least 8 decimal digits, so the location can be stored with
maximum precision for later enabling auto tagging. The coordinates must be in accordance
with ISO 6709. The accuracy standard must be in accordance with FGDC-STD-007.

File Format: All digital deliverables will be in Windows compatible file format.

Point cloud providers shall deliver registered point cloud data Scan data registration to the
site/target monuments/control points, for enabling integration with existing data models and
geo referencing capabilities. A report of the site/target monuments/control points containing
the location, coordinates, EPSG Code (when georeferenced) and Transverse or Survey Grade GPS
calculations must be provided.

Naming convention: Vendor will deliver imagery and E57 or LAS scan files in its native format
with the same root name in the following format. The below example is for an onshore oil field
with multiple Pad Locations but can be adapted for other scenarios/assets:

• [I|E Interior or Exterior] _[PAD identifier Alphanumeric string]_[Location identifier Alpha-

numeric string]_[YYMMDD zero padded date of scan]_[NNN zero padded provider
sequence number]_[C|I|S Color, Intensity, Scan]
• “_” is a reserved field separator character not to be otherwise used in the name.
• Interior is typically located in controlled environment and in the case of scan data, has
been registered to a local interior survey control network.
• Exterior is typically exposed to weather and is registered to an exterior survey control
network.
• PAD identifier e.g. Drill site/Facility identifier – short unique common name.

Open Asset Digital Twin Working Group • 14

 OADT Reality Capture Guidelines• April 2025

• Location Skid/Module/Section/Unit/Rack identifier – short unique name using com-

mon facility terminology that identifies in what area on the PAD the scan was taken.
• Date of scan in zero padded YYMMDD format, where YY is last two digits of the year,
MM is the two digits of the month, and DD the two digits of the day in the month.
• Vendor selected Zero padded sequence number NNN, so file names are unique for same
area-same day scans. It is preferred that values be sequential without gaps, but not re-
quired. Values may duplicate with other days in the same area.
• Type of file e.g. C for color imagery, I for intensity/gray scale imagery, S for scan data.
• Example Laser scan, color imagery, and gray scale (Intensity) files for one capture
of an Interior scan at Alpine facility, CD5C module, taken June 23, 2020.
• Laser Scan file: I_ALP1_CD5C_200623_001_S.e57
• RGB image file: I_ALP1_CD5C_200623_001_C.jpg
• Intensity image file: I_ALP1_CD5C_200623_001_I.tif
• HDR image file: I_ALP1_CD5C_200623_001_C.exr
• Artifacts of a scan position can optionally be grouped in a folder with the noted
root name e.g. folder name of “I_ALP1_CD5C_200623_001”.

Example: Metadata for Reality Capture Project

Open Asset Digital Twin Working Group • 15

 OADT Reality Capture Guidelines • April 2025

Survey:
• Digital Survey report showing accuracies of the control points used/surveyed and the overall
network.
• Format shall include four titled columns of:
• Point Identifier
• Easting
• Northing
• Elevation
• Measurement units e.g. feet, yards, meters, etc. shall be prominently noted in the co-
ordinate column name e.g. “Easting – Feet.”

Scan Data
Data from each processed scan position will be delivered as one scan per file of ASTM E2807
aka E57 file format.

3D Imaging Data Exchange | ASTM

https://www.astm.org/news/3d-imaging-data-exchange-mj11

Scan Imagery
Imagery - 360 spherical / equirectangular panoramic images in both Intensity and Color forms
as captured during the scanning process.

• Intensity will be the measure of the laser return signal for each measured point.
• Color imagery will have sufficient dynamic range to reasonably view both shadowed and
brightly lit areas in the same view.
• All processed Imagery files should be delivered in .JPG or .TIF formats and shall fully repre-
sent the resolution of the capturing device without transformation e.g. not as mapped to scan
points or having been thinned.
• Images exported in .JPG format will be compressed no lower than 90% quality setting.
• Deliverables should include both the raw and processed imagery.
• HDR source imagery (the individual stepped ISO pictures) shall be delivered in. EXR (pre-
ferred), .HDR file format, or as a sequence of .TIF images located in a folder named for that
scan position, along with the integrated processed final HDR image file.

Accuracy

The Reality Capture Service Provider is solely responsible for the selection and operation of all
tools, processing of results, and correctness of all deliverables.

Open Asset Digital Twin Working Group • 16

 OADT Reality Capture Guidelines• April 2025

The Laser Scan Service Provider must always use calibrated professional-grade scanning equip-
ment with appropriate quality settings that ensure capturing the specified scope at or better than
the Engineering Grade quality levels unless otherwise stated in the statement of work.

Using one or more of the listed scanning devices does not guarantee data will meet the specifi-
cations stated herein and the Service Provider is solely responsible for delivering the artifacts of
the specified quality. Using a quality laser scanner does not guarantee the quality of results, but
it does enable the possibility of quality when used in concert with appropriate processes and
controls.

Scan data should be captured with all assets being scanned in a steady-state normal operating
condition and if that is not feasible, clearly document the state of the assets captured. A signifi-
cant movement of physical plant surfaces can occur between quiescent and normal operational
states. These differences will cause distortion and inaccuracies if scan data from one state is
merged with data captured in a differing state.

Registration between scans and registration of scans to plant or the local coordinate system will
be performed using surveyed control targets unless otherwise stated herein.

Accuracy: Engineering Grade (Default)

• The required accuracy unless otherwise stated in the statement of work.

• Intended to provide sufficient accuracy to support the fabrication and routing of most assets.
• When using scanner settings for reality capture the scanner will have a maximum error
bound of no more than 0.08 inch (2mm) @ 32.8 ft (10m) for range accuracy, and 0.02 inch (0.5
mm) @ 32.8 ft (10m) for range noise.
• Minimal errant measurements at edge boundaries regardless of background; these may be
removed during scan post-processing.
• Capable of measurement spacing (aka Density) of ≤ 3.2mm @ 10m in a single scan; equivalent
to ~19,635 measurements at the horizontal. Scanning at this density is not required, however
the capability to scan at this density or higher is critical when scan positions are limited and
assets to be captured are at a distance e.g. exhaust stacks, at height platforms, or close in is
not possible e.g. flare pits.

Scan-to-Control Registration

• Each laser scan station will be tied to the local control network in such a manner that the
registration fitting errors between any scan and the local control network do not exceed
±0.125 inches (±3.2mm).

Open Asset Digital Twin Working Group • 17

 OADT Reality Capture Guidelines • April 2025

• The laser scan service provider shall align each laser scan station to neighboring (those tied
to the same survey local control network) scan stations in such a manner that the registration
fitting errors between any scan and the local control network do not exceed ±0.125 inches
(±3.2mm).

Scan-to-Scan Registration

Each laser scan station will be registered with other scan stations on the same local control net-
work in such a manner that the registration fitting errors between any two scans do not exceed
±0.125 inches (±3.2mm).

Tolerances

The horizontal and vertical tolerances of any measurement taken within any registered point
cloud dataset, that are all aligned to the same survey local control, shall be within ±0.125 inches
(±3.2mm) of the corresponding surveyed physical measurement taken in the field under similar
operating conditions as when scanned.

Accuracy: Navigational Grade

• For capturing areas that only require situational awareness. The dimensional accuracy is sub-
stantially less accurate than Engineering Grade, but still has the coverage and density to
support many workflows and use cases.
• Uses scanning equipment, typically handheld or mobile, that is specified to have a maximum
error bound of no more than ±0.60 inches (±15mm) between any two points located within
10m of each other.
• Requires rigor by the Service Provider in establishing survey control to enable adequate di-
mensional placement of captured points and to substantiate the overall error bound.

Scan-to-Control Registration

Each scan path will be tied to the local control network in such a manner to ensure accuracy levels
are maintained throughout the captured area.

Scan-to-Scan Registration

The use of scan-to-scan registration with mobile scanning is discouraged. However, it is allowa-
ble if the Service Provider can substantiate through independent means that the resulting data
meets the required quality levels.

Open Asset Digital Twin Working Group • 18

 OADT Reality Capture Guidelines• April 2025

Tolerances

The horizontal and vertical tolerances of any measurement taken within any registered point
cloud dataset shall be within ±0.60 inches (±15mm) of the corresponding surveyed physical meas-
urement taken in the field under similar operating conditions as when scanned.

Survey Control to Plant Coordinates

The objective is to have coordinates of scan data aligned with documented plant coordinates of
production assets. Survey Control coordinates will be established from Plant Coordinates as doc-
umented by the Operator. Documented plant coordinates of production Assets are considered
authoritative.

Horizontal Survey Control

The Scanning Contractor will ensure the accuracy (uncertainty) of horizontal measurements for
the purpose of establishing horizontal survey control. Surveyed control points shall meet the
requirements of 2-Millimeter @ 95% confidence as defined by the “FGDC-STD-007.2-1998 Geo-
spatial Positioning Accuracy Standards Part 2: Standards for Geodetic Networks, FGDC Table
2.1.”

Vertical Survey Control

The Scanning Contractor will ensure the accuracy (uncertainty) of vertical measurements for the
purpose of establishing vertical survey control. Surveyed control points shall meet the require-
ments of 2-Millimeter @ 95% confidence as defined by the “FGDC-STD-007.2-1998 Geospatial
Positioning Accuracy Standards Part 2: Standards for Geodetic Networks, FGDC Table 2.1.”

Scan Density and Coverage

To qualify as Scan Coverage in delivered scan data, an area of adjacent scan points possibly from
different scans, must meet the Scan Accuracy requirements and have a minimum Scan Density
of 16 measurements per inch (6.299 measurements per cm) as measured in two perpendicular
directions.

Scan Coverage

The objective is for 100% of every object surface to be laser scanned at the required minimum
density to present a complete image of all visible surfaces within the target volume. It is under-
stood that this is not realistically achievable for all except the most simplistic environments.

Open Asset Digital Twin Working Group • 19

 OADT Reality Capture Guidelines • April 2025

To qualify as Scan Coverage, adjacent scan points possibly from different scans, must meet the
Scan Accuracy and Scan Density requirements.

Target coverage of items listed below is understood to be minimums that may still be further
constrained by the realities in the field. The ability to consistently achieve no less than the mini-
mum coverage is a positive indication of vendor skills and initiative.

Piping – used to determine pipe component type, size, rating, dimensions and layout

• 100% of significant markings e.g. pipe identifier or inspection point labels

• 40% of each flange lip
• 40% of each Tee, Lateral, or Cross component
• 40% of any directional change e.g. bend or elbow
• 40% of each reducer
• 30% of each tap connection point e.g. vents, drain, instrumentation connection
• 30% of each fitting
• 30% of each intermediate attachment point e.g. hangers
• 20% of general surface between end points or directional change points

Vessels and Tanks

• 100% of significant markings e.g. vessel identifier

• 40% of each Nozzle
• 40% of each external process connection
• 75% of each human access point
• 40% of each surface area that changes orientation relative to its surrounding
• 60% of each attachment point e.g. footing, lift point
• 40% of general surface.

Equipment

• 100% of significant markings e.g. equipment or equipment component identifier

• 50% of general surface
• 60% of each attachment point e.g. footing, lift point
• 40% of each external piping connection.

Structural

• 100% of significant markings e.g. identifiers

• 50% of each attachment or contact points
• 50% of directional changes

Open Asset Digital Twin Working Group • 20

 OADT Reality Capture Guidelines• April 2025

• 30% of general surface.

Instrumentation

• 100% of significant markings e.g. instrument identifier

• 50% of general surface area

Other not declared elsewhere

• 100% of significant markings e.g. identifiers

• 60% of each attachment or contact point
• 75% of each human access point
• 60% of each attachment point
• 40% of general surface.

Target Group and Responsibility

Those responsible for delivery of Reality Capture artifacts representing Operator Assets.

Typical Project Team Members

include, at a minimum, a facility operator and resident survey contractors, but may also include
planner/scheduler, HSE, maintenance techs, or management depending on the scale and scope.
Larger projects should involve the following resources

• Operator:
• Facility operator
• Resident Survey Contractor
• Planner/Scheduler – planning, scheduling, and coordination of scan timing with other work
• HSE – Verification that the timing and scan location are safe for performing the work.
• Maintenance techs
• Permitting – Issuance of required work permits, defeat of safety systems/energy isolations
as required
• Service Provider
• Capture Lead
• Capture Team

Depending on the scale and scope of the area to be captured, one person may perform more than
one role, and/or the number of people in each role may differ.

Open Asset Digital Twin Working Group • 21

 OADT Reality Capture Guidelines • April 2025

Photogrammetry

Fundamental Principles
Photogrammetry is based on the principle of triangulation. By taking photographs from multiple
viewpoints and using software to identify common points across these images, it's possible to
triangulate the coordinates of these points in three-dimensional space. This process requires care-
ful alignment of photos, ensuring that each point is captured from at least two different locations
to intersect lines of sight mathematically and produce accurate spatial data.

Capture Workflow
Planning: Proper planning involves analysing the shooting environment and determining opti-
mal lighting and angles. This ensures comprehensive coverage and sufficient overlap between
images for accurate point matching.

Capturing Images:

• Utilise stable platforms such as tripods or drone gimbals to prevent blurring and ensure con-
sistent overlap and coverage.
• Capture a series of overlapping images (60-80% recommended) from various angles to cover
all aspects of the subject.
Processing:

• Image Alignment: Software processes the images to find common points and estimate cam-
era positions.
• Point Cloud Creation: From the aligned images, a dense point cloud is generated, represent-
ing the object’s external surfaces as thousands or even millions of points.
• Mesh Generation: The point cloud is transformed into a 3D mesh, which involves creating a
network of polygons that approximate the object’s surface.
• Texturing: The 3D model is overlaid with textures derived from the original photographs,
providing realistic details and colours.

Output and Analysis: The final model can be used for various analytical purposes, rendered for
visual presentations, or exported for further processing in other digital tools.

Capture Guidance
Data Management: Effective data management is crucial in photogrammetry to ensure that the
data collected is accurate, secure, and readily available for analysis and application. The follow-
ing are some key principles of data management for photogrammetry data:

Open Asset Digital Twin Working Group • 22

 OADT Reality Capture Guidelines• April 2025

Data Collection and Integrity:

• Standardization: Establish standard protocols for data collection, including camera

settings, angles, overlaps, and environmental conditions, to ensure consistency across
different projects.
• Validation: Implement processes to verify the accuracy and completeness of the cap-
tured images immediately after collection. This might include checking for sufficient
overlap, focus, and exposure to prevent data loss and rework.

Data Storage:

• Security: Secure storage solutions are essential to protect sensitive or proprietary

data from unauthorized access or data breaches. This involves using encrypted stor-
age and secure access protocols.
• Redundancy: Employ redundancy systems, such as RAID configurations or cloud
backups, to prevent data loss due to hardware failure.

Data Processing:

• Efficiency: Optimise the data processing workflow to handle large volumes of data
efficiently. This can involve using powerful processing software and hardware, and
potentially offloading some of the heavy processing to cloud services.
• Scalability: Ensure that data processing systems can scale up to handle increases in
data volume without significant slowdowns or degradation in performance.

Data Analysis:

• Accessibility: Make data easily accessible to authorized users, with tools and inter-
faces that support efficient analysis and visualization of results.
• Interoperability: Use formats and standards that ensure data can be easily shared
and used across different systems and software, enhancing collaboration and utility
across various applications.

Data Archiving and Retention:

• Long-term Storage: Establish protocols for archiving data that is not immediately
needed but may be valuable for future projects or historical records.
• Data Lifecycle Management: Implement policies for the lifecycle of photogrammetry
data, from creation to disposal, including when and how to safely and legally dispose
of data that is no longer needed.

Open Asset Digital Twin Working Group • 23

 OADT Reality Capture Guidelines • April 2025

Documentation and Compliance:

• Metadata: Maintain thorough documentation of data, including metadata that details

the data’s origin, parameters, and any alterations made during processing.
• Legal and Regulatory Compliance: Adhere to legal and industry-specific regulations
regarding data handling, privacy, and archival, which can vary significantly depend-
ing on the application area and geographical location.

Deliverables

Data Content

Photogrammetry providers shall deliver high-resolution images per capture position

Recommended imagery specs should follow use cases. The raw files won’t be stored since the
panoramic pictures are widely accepted in photogrammetry tools and require less storage space.
A report of the site/target monuments/control points containing the location, coordinates, EPSG
Code (when georeferenced) and Transverse or Survey Grade GPS calculations must be provided.

Photogrammetry providers shall deliver the metadata for each image file to allow traceability
and cross-referencing the capture positions. Every nomenclature used must follow the taxonomy
references based on industry standards.

Data Format

The format for the high-resolution images is JPEG or PNG. Both are low compression formats
used in digital photography, widely used for high-quality pictures and it is supported by all
browsers and image editing software. These formats are standardized by ISO/IEC 10918,
ISO/IEC 15444 and ISO/IEC 15948.

Pictures files shall preferably be available in open-source formats

• Images: GIF, JPEG, PNG, SVG, BMP, TIFF

Open Asset Digital Twin Working Group • 24

 OADT Reality Capture Guidelines• April 2025

Example of Photogrammetry Metadata for a reality capture

Accuracy

Several factors affect the accuracy of photogrammetric models:

• Camera Quality and Calibration: High-resolution cameras with minimal lens distor-
tion are critical. Calibration helps correct any inherent optical distortions.

• Image Quality and Overlap: Ensuring uniform lighting and sufficient overlap between
images is crucial for accurate point matching and triangulation.

• Software Capabilities: Advanced software features can greatly enhance point detec-
tion, matching accuracy, and model resolution.

Examples of Tools and Technologies

• Cameras: High-quality DSLRs or mirrorless cameras, such as the Nikon D850 or Canon
EOS 5D, are preferred for their high resolution and robustness. For larger areas or inac-
cessible locations, drones like the DJI Phantom 4 are used.

- Basic DSLR Cameras (e.g., Canon EOS Rebel T7):

• Accuracy: 3-5 mm

• Range of Capture: Suitable for small to medium-sized projects.

- Advanced DSLRs or Mirrorless Cameras (e.g., Nikon D7500):

• Accuracy: 1-3 mm

• Range of Capture: Effective for medium to large projects.

Open Asset Digital Twin Working Group • 25

 OADT Reality Capture Guidelines • April 2025

Consumer Drones (e.g., DJI Mavic Air 2):

• Accuracy: 1-5 mm, varies with altitude and sensor quality.

• Range of Capture: Capable of covering large areas, ideal for landscapes or large struc-
tures.

High-Resolution DSLRs or Mirrorless Cameras (e.g., Nikon D850, Canon EOS 5D Mark IV):
• Accuracy: Less than 1 mm

• Range of Capture: Extensive, with capabilities for detailed capture of large and com-
plex scenes.

Professional Drones (e.g., DJI Phantom 4 RTK):

• Accuracy: Around 1-2 cm or better with RTK (Real-Time Kinematic) modules.

• Range of Capture: Extensive, can cover large geographic areas with precise geotagging.

Definitions

Virtual Digital Twin: A "virtual object digital twin" refers to a computer-generated, virtual rep-
resentation of a physical object that closely mirrors its real-world counterpart, allowing for
detailed analysis, simulation, and monitoring of its behavior using real-time data from sensors,
essentially creating a "twin" of the physical object in the digital realm.

Photogrammetry: Photogrammetry is the science and technology of obtaining reliable infor-

mation about physical objects and the environment through the process of recording, measuring,
and interpreting photographic images and patterns of electromagnetic radiant imagery and other
phenomena.

Laser Scanning: Laser scanning is a non-destructive method using a laser beam to obtain meas-
urements of remote surfaces, typically in a grid-like fashion. This process is commonly known as
Lidar (Light Detection and Ranging)

SLAM: Simultaneous Localization and Mapping. SLAM is a fundamental technology in robotics

that enables robots and other autonomous systems to build a map of an unknown environment
and determine their own location within that map simultaneously

Aerial Survey: An aerial survey is the process of collecting data and information about a large
area of land or terrain from an aerial perspective, typically using aircraft or drones equipped with
specialized sensors and cameras. 1

Open Asset Digital Twin Working Group • 26

 OADT Reality Capture Guidelines• April 2025

Qualified Laser Scanner: A device that the manufacturer states is designed to meet the specifi-
cations noted herein when capturing surface measurements of assets using common industry
laser scanning processes and workflows. (Refer to section 4)

Qualified Laser Scan Service Provider: A service provider that attests to having and demon-
strates the necessary skills, processes, experience, and equipment to safely capture and deliver
laser scan data artifacts that consistently meet the quality and scope objectives of this specifica-
tion and are fit for the contracted use cases. (Refer to section 7)

Scan Device Accuracy: The ability of a laser-scanning device to consistently return a measure-
ment to an expected surface within a specified error bound.

Colored Scan Data: Laser scan values are inherently monochromatic. To colorize the values, a
photographic image is taken from the same position as the laser scan and is aligned with the scan
data. The pixel(s) from this photo image are used to derive an RGB value for each scan point
resulting in colorized scan data. Note the RGB data can be visually misleading, even erroneous
due to the mapping process e.g. colorized scan points display a surface edge, but the underlying
scan data shows the edge in a different position.

• Colorized scan points will include the laser scan return signal intensity (≥256 values of bright-
ness) and from separate photography imagery, a generated derived color (≥256 values x3 Red
Green Blue) for that point. Good quality visualization tools will allow the user to display scan
data as Intensity or RGB.

Plant Coordinate System: A flat (non-geodesy) 3D Cartesian coordinate system referenced from
local survey benchmarks utilized within plant drawings to detail the relative X, Y, and Z (Eleva-
tion) position of key asset points throughout the facility site. Once established, the documented
key coordinates of plant assets become the authoritative reference for plant control.

Local Control Network: A subnet of surveyed control points that are initially set to the overall
plant coordinate network and are expected to drift over time largely in unison e.g. the control
points within a module. The Local Control Network is used to establish and anchor the coordi-
nates of scan data within this area. A local control network is highly recommended as greatly
simplifies merging multiple laser scans and supports integration across multiple different data
sets.

Scan Accuracy to local coordinates. The maximum expected difference (error bound) between
the local coordinate of a point in the delivered scan data and the first-order class 1 surveyed local
coordinate (1:100,000) of the same physical point under the same environmental conditions.

Open Asset Digital Twin Working Group • 27

 OADT Reality Capture Guidelines • April 2025

Scan Measurement Accuracy: The maximum expected measurement difference between two
scanned points versus the same measurement using survey equipment between the same two
physical points under the same environmental conditions, where:

• Both points are in the same scan

• Both points are in the same interior survey Local Control Network, but different scans.
• Both points are in the same exterior survey Local Control Network, but different scans.
• Both points are in a different survey local control networks and different scans, but still
within the same plant coordinate system.

Raw Scan Density: The inverse of the linear distance between a scan point and the furthest away
adjacent Cartesian neighbor scan point on the same planar surface in a single scan, where such
surface is largely perpendicular to the view of the scanner. Effectively the number of measure-
ments on a surface per unit length along a scan axis from a single scan sweep and the distance
between consecutive scan sweeps of the same scan on the same surface. At a specific measure-
ment frequency, the greater the distance between the scanner and the surface measured, the
larger the spacing between measurements and therefore a lower density of measurements on the
same surface area.

Scan Density The inverse of the average linear distance in the delivered scan data between adja-
cent scan points that meet the accuracy requirements, possibly from different scans, and are
located on the same planar surface. Scan Density delivered is generally in the aggregate (a surface
captured by two or more scan positions) which is not equivalent in quality to the same Raw Scan
Density. Aggregate scan points will result in a moiré of scan points that are never perfectly inter-
leaved, and the points from different scan positions will have different error bounds.

Raw Laser Scan Data: The digital data output from a laser-scanning device before any post-scan
processing.

Raw Image Data: An image file containing the intensity/brightness values from the laser scan
return signal of each measured point. An image file containing the RGB imagery. For RGB, the
imagery is from a digital camera either intrinsic with or separate from the laser scanner before
any post-image processing. This is often referred to as Panoramic Imagery or Panos. Most mod-
ern terrestrial scanners will simultaneously acquire both Laser Scans and Colored Panoramic
Images.

HDR Imagery: HDR, High Dynamic Range, is a technique for widening the exposure range so
details in a picture are not lost in a combination of bright and dark areas. This is accomplished
by taking a sequence of pictures of varying exposure to better capture the brightest or darkest
areas, then merging the best aspects of each image to achieve a single image that shows details

Open Asset Digital Twin Working Group • 28

 OADT Reality Capture Guidelines• April 2025

in both the bright and dark areas. HDR can provide superior RGB imagery for colorizing the laser
scan points.

Scan Registration: The act of aligning and placing captured data into a common coordinate sys-
tem is colloquially called registration. Registration is generally accomplished using cloud-to-
cloud and/or surveyed targets.

Cloud-to-Cloud Registration, aka Cloud Registration: A process, often with some automation,
that aligns the coordinates of scans by matching common features seen across different scan sets.

Scan Target: A geometric shape that can be easily recognized by a laser scanner to derive a com-
mon reference point between scans and, if surveyed, constrain / anchor the scan data to surveyed
coordinates. Common target shapes are a checkerboard image or a reflective sphere of calibrated
size.

Target Registration: A process, often with some automation, that aligns the coordinates of scans
by matching the view of common Scan Targets seen by different scans. The targets are distributed
throughout the scanning area before scanning. If permanently mounted, the targets enable future
scan data to be aligned with historical scan data.

Survey Control Target: A Scan Target that has been surveyed, generally using a Total Station, to
identify its location in plant coordinates and thereby both enable the alignment of scans that view
this target and constrain the position of the scanned points for better alignment of each respective
scan.

Reality Capture Deliverable: Deliverable to Operator of processed scan measurement data, im-
agery artifacts, and reports that the service provider has processed, validated, and attests to meet
the required specifications. Contracting scan companies are encouraged to upload final datasets
to the Owner Operators preferred system of record.

Capture Area Scope: The contract specified boundaries of the area volume that has been desig-
nated for capture. Incidental scanning of areas outside of the requested volume while achieving
the request scope but that otherwise meet Operator quality specifications must be delivered with
the requested scope. A scan or imagery that contains elements of both in-scope and out-of-scope
data shall not reduce or clip the out-of-scope data in any manner.

Survey Reference Monument: The survey fixed physical control point that defines one or more
elements of the northing, easting, and elevation in a known coordinate system. Although in-
tended to be stable, always cross-check the position of a monument with other known points as
it may have experienced substantial movement, especially in cold climates where frost heave can
move the ground (Alaska, Canada etc.).

Open Asset Digital Twin Working Group • 29

 OADT Reality Capture Guidelines • April 2025

Reference Survey Control: Reference monuments that identify the northing, easting, and eleva-
tion of the plant or local coordinate system. Delivered laser scan data will have been transformed
and aligned to the plant coordinate system using generally accepted survey methods.

New/Bulk Capture: Utilizes equipment and procedures to capture the entirety of an area to the
quality and coverage specifications. New Capture is appropriate for areas not previously cap-
tured or previously captured areas that have experienced physical change over a large area.
Special care must be taken to establish accurate permanent local survey control if not already
present. Unlike Update Capture, the resulting artifacts of New Capture are not typically melded
with previous captures.

Update Capture: Uses procedures and equipment to capture visible changes within a defined
area that have occurred since the previous capture. Unlike New Capture, Update Capture scans
of a physically changed area must accurately align to the local control to properly meld with all
previously captured data in the same area. (See “Scan Registration”)

Significant Markings: Markings that are commonly used to visually communicate from a dis-
tance such as notation or signage symbols that are at least 0.75 inches (48 pt) in height and are
positioned to be visible from typical worker locations.

Acronym Reference:
AI Artificial Intelligence IIoT Industrial Internet of Things
AR Augmented Reality IoT Internet of Things
CAPEX Capital Expenditure IT Information Technology
COTS Commercial Off the Shelf OPEX Operational Expenditure
CWE Collaborative Work Environment OT Operational Technology
IIS Industrial Internet System SLA Service Level Agreement

Open Asset Digital Twin Working Group • 30

 OADT Reality Capture Guidelines• April 2025

Founded in 1986, ARC Advisory Group is the leading technology research and advisory firm for industry and infrastructure.
ARC stands apart due to our in-depth coverage of information technologies (IT), operational technologies (OT), engineering
technologies (ET) and associated business trends. Our analysts and consultants have the industry knowledge and first-
hand experience to help our clients find the best answers to the complex business issues facing organizations today. We
provide technology supplier clients with strategic market research and help end user clients develop appropriate adoption
strategies and evaluate and select the best technology solutions for their needs.

ARC Advisory Group, Three Allied Drive, Dedham, MA 02026 USA • 781-471-1000 • www.arcweb.com

Open Asset Digital Twin Working Group • 31

3 ALLIED DRIVE DEDHAM, MA 02026 USA 781-471-1000

USA | GERMANY | JAPAN | KOREA | CHINA | INDIA | SINGAPORE | BAHRAIN & UAE | BRAZIL