Comments provided for TCP…

Comments provided for TCP 000-0047 are applicable.

Thank you for the opportunity to provide comments re. amendment to OPSS 510 specification. We believe that, Implementation of 3D milling technology is a very important technical step, that would not only benefit the road industry, but our society in general for providing safety, the comfort and minimizing CO2. Per our understanding following discussions with several dealers, 3D milling is a technological process that allows milling a road according to a 3D model. This 3D model is prepared in advance in the office, all the problems of the construction are solved and the model is approved. Then there are no unpleasant surprises for the workers on the construction site. The quality of the milling is primarily determined by the quality of this 3D digital design model (DDM), and the quality of the DDM is limited primarily by the quality of the Road Surface Survey. We have attached comments summarizing our understanding of the 510 proposal, including comments, questions, and general feedback bothering on definition of the tolerances of the accuracy of each step and the inspection procedure, as well as instructions to designers. Summarily, we believe that, what kind of software is used for 3D modeling of DDM or what hardware (AGM -Automated Machine Guidance) is mounted on the milling machines for 3D milling control is not the most important thing. What is crucial is the accuracy of the input data => the road surface survey before milling. Bad inputs mean bad results. This is also the conclusion of the U.S. Department of Transportation study, "UTILIZING 3D DIGITAL DESIGN DATA IN HIGHWAY CONSTRUCTION - CASE STUDIES", which concludes on Page 178: "It was surprising to find that the issues with design data-such as inconsistency with the design intent in the plans-took time to resolve, but had a minor impact on construction outcomes. The construction issues were a result of the unreliable original ground survey that the design was based on." We encourage the Ministry to take at least a cursory look at this study, while considering the detailed comments and concerns and questions we have provided, thank you. https://www.fhwa.dot.gov/construction/3d/hif17027.pdf.

Our Understanding:
Our understanding of the amendment to OPSS 510 is that, the specification has been divided into several steps the contractor is to complete, while also establishing control mechanisms:
1. Road Surface Survey
a. Establishment and surveying of Survey Registration Points (by digital leveling)
b. High-density survey (e.g., LiDAR - 3D scanning)
2. Digital Road Surface Model (DRSM)
3. Digital Design Model (DDM)
4. Automated Machine Guidance (AGM)
These individual steps build on each other and each adds some error to the overall process. The total sum of the errors must be within acceptable limits for the result of the repair process to meet the required parameters for IRI, cross-fall, surface drainage, match the elevation of fixed appurtenances including curbs, manholes, and others. Other requirements are, for example, for future follow-up design work, so the entire DDM must be tied to the "Owner's geodetic control and benchmarks" (as defined in 510.07.06.04.03). These points are fixed in time and therefore can be used in the future, for example, to delineate bridge structural elements.

Primary comments bothering on definition of the tolerances of the accuracy of each step and the inspection procedure:

1) 510.07.06.04.03: The definition of the vertical accuracy of Survey Registration as tolerance of +/- 4mm with respect to existing geodetic control is not sufficient. This definition would allow 8mm vertical difference between the adjacent registration points.
We propose this standard and commonly used definition among surveyors has to be added to existing definition: The accuracy of the differential leveling must be demonstrated by calculating a closed leveling traverse that must not exceed 8 mm per 1 km or/and +/- 2mm difference between adjacent registration points.
The selection of the locations (STA) where these quality "control cross section" measurements will be taken must be predefined and must not depend on the contractor's choice.
We suggest, for example, a midpoint between the survey registration points with a position tolerance STA +/- 10m; where maximum errors are assumed.
“An automatic machine guidance (AMG) system shall be installed on the milling equipment used for the work of removal of partial depth asphalt pavement removal. The AMG system shall be capable of precise three-dimensional control of equipment movement using satellite and local referencing.” Does this mean the need for base stations or something else?
“The AMG system and digital machine control file shall automatically control the milling equipment such that the existing asphalt pavement is partially removed over its entire surface to match the vertical dimension of the DDM milled surface to within a ±5 mm tolerance.”
Does MTO have data to confirm this is constructible?

“The Contract Administrator will carry out total station measurements of the milled surface to verify the ±5 mm tolerance is met.“
If the QC and QA measurements differ, will a referee system be available to the contractor to rectify?
“The survey registration points shall be referenced to the Owner’s geodetic control and benchmarks at the limits of the removal of asphalt pavement, partial depth. The geodetic control and benchmarks used shall be sufficiently distant from construction operations to be protected from disturbance.”
Will geodetic information be available, practical, and readily accessible for all MTO contracts where this specification is employed?
“A high-density survey of the existing asphalt pavement surface area using high accuracy methods shall be used to collect a minimum of 1000 measured points / m2 or of higher density as required to meet the standard deviation requirements specified herein. Each point shall be measured in three dimensions. The high-density survey shall be registered to the surveyed registration points.”
This is extremely high and will consume a huge amount of data with no advantage. A milling machine cannot change and react to even 1 point per m2. Why was this resolution selected? Check with manufacturers for verification.
2) 510.07.06.04.04: A check of the accuracy of the DRSM that is made by measuring the "control cross section" must meet a better criterion than the currently defined 8mm standard deviation of elevation differences, as this could lead to an error in slope of 0.7% or more out of the figure. Standard deviation of 8mm could result in elevation differences up to 24mm – please see attached drawing.
We suggest 3 mm standard deviation which is achievable with reasonable effort and reduce implementing of huge error from all beginning of entire process.
“The standard deviation of the calculated elevation differences for all the points at each cross-section location, z, shall not exceed 8 mm.”
Does this coincide with the 4 mm tolerance in 510.07.06.04.03?
“A DRSM meeting the standard deviation accuracy requirement shall be used to create a digital design model (DDM) of the milled surface resulting from the removal of asphalt removal, partial depth work, and the subsequent layer(s) of asphalt materials to be placed.”

3) 510.07.06.04.05: The Automated Machine Guidance technology for 3D milling in section should also allow manual entry of milling depth values (for example, values to be displayed on the device tablet based on GPS position).
If the milling operator will be able to meet the accuracy criteria by such a procedure there is no reason why this procedure shouldn´t be allowed. This will also allow the use of older milling machines and this paragraph will not discriminate against older and still high quality milling machines.
The final definition for checking the correct milling depths (510.07.06.04.05) according to the DDM, which must meet +/-5mm, should be better described and should include checking the milled slope to match the design slope according to the DDM at that location. The DDM check of cross slopes is very important because it involves not only checking the AMG function and milling machine calibration, but also checking that the DRSM has been well measured and that the slope correction has been correctly designed in DDM.
Merely checking, the depths is just a relative reading between the DRSM and the DDM and would not reveal an erroneous cross-fall.

Other Areas of Concern with Questions and General Feedback:
4) 510.04.02.01: “The DSRM shall be digitally sealed and signed by an Ontario Land Surveyor or Engineer with specialist training in Geomatics”.
Can this not be sealed and signed by a P. Eng. instead?
Will MTO be developing and providing contractors a list of licensed Ontario Land Surveyors qualified for this work who are acceptable to the MTO?
Sufficient time will need to be allotted by MTO for this service.
5) 510.07.06.04.02: “After partial depth removal, the gap between the top of milled surface and the bottom of a 3 m straightedge placed anywhere in any direction on the milled surface shall not exceed 6 mm.”
Top of milled groove, correct? What about the distance between the bottom and top of milled groove?
Except at termination joints? Is this measurement to the top (ridge) of the milled surface or the bottom of the milled surface (valley)?
“The surface remaining after removal shall have a constant and continuous crossfall matching the design milled surface crossfall. The milled surface shall have an even texture and be free of significantly different grooves and ridges in all directions.”
To what tolerance?

6) The "3D Guidance" of the machine language needs revisited. The guidance for the machine should not be dictated by method but by accuracy. Machine control systems today provide "variable depth and slope", but don't steer the machine. If the intention of the spec is to say that the machine is controlled both vertically and horizontally, we think this is unachievable.
As a point of clarity, the depth and slope are controlled automatically, the horizontal position of the mill is still up to the operator and the design will reflect the mill position accordingly.
We suggest stating the expected accuracy of the system in both the horizontal and vertical and allow the contractor to select the guidance that will meet or exceed that specification.

7) It doesn’t seem the spec., considers the road scanning requirement of a tripod mounted total station, why?
Perhaps this should be spec'd as an accuracy and not a method. If mobile scanning can meet or exceed the intended accuracy (both in terms of horizontal and vertical accuracy, but also in point density) then it should also be allowed, right?

8) Will the MTO be performing preliminary design prior to tender to ensure the final design is achievable?

9) How will MTO determine whether the contract will be a contractor-provided survey or MTO provided survey?

10) What type of contracts will MTO be utilizing this specification on? Ie. freeway, 2 lane rural, etc. Will line of sight to geodetic benchmarks be consistently available?

Finally, re. “INSTRUCTIONS TO DESIGNERS”
Warrant: “Sufficient pavement depth to limit risk of “punch-through” to granular base by variable depth / optimized milling”
This is very critical! There have been many historical contracts where the existing HMA thickness was not sufficient to accommodate the specified milling depth.
Will the MTO perform pre-engineering assessments to assure sufficient pavement depth exists?