I'm attempting to find digestible information about EC Schemas. I'd like to help a friend who is a worldwide BIM manager for a well-known engineering company. Can I find anything on the Bentley Systems website? Nada, nothing, nic w ogóle, niente di niente, nichts, rien du tout.
By 'digestible' I mean resources that use everyday English to describe EC Schemas, what they are used for and their strategic potential. As a developer of software for MicroStation, I've been aware for many years of EC Schemas and the technology's potential in Bentley Systems' base platform. However, as a strategic technology there's nothing on offer for the man-in-the-street wanting to know more. How do I convince someone that MicroStation and its vertical apps. are a good path to take towards BIM and data exchange? How do I counter arguments along the lines of 'Revit already does that'? Where can someone find examples of core technology EC Schemas emerging in vertical apps. such as AECOSim?
Jon Summers said: How do I convince someone that MicroStation and its vertical apps. are a good path to take towards BIM and data exchange?
Well, I don't even think that this is something that most big international engineering firms even consider to be a goal. The majority of the engineers in the organization at the top of the hierarchy with strategic decision / budget making means are fundamentally 'pen and paper... I manage people and contracts, not calculate stuff' guys, with very little understanding of digital processes. They tend to take a 'let the market sort it out' attitude which is bad for Bentley.
At one un-named major UK outfit, it was clear from the internal forums, that Autodesk and others had far more mind share and training material and offers for external seminars and help, compared to Bentley. The projects had a laissez faire approach to software choice, where the engineers were allowed to use whatever they were most comfortable with.. with recent grads firmly more familiar with AD products... even within the infrastructure group, where Bentley products were once dominant but now look stale and buggy, with minimal support from the vendor, who will only change/customise things when paid large sums. I guess the lack of third party software devs make £1k+/day charges common, but way out of what most engineers would be authorised to spend.
The interoperability problems also tend to swept under the carpet by the project engineers because they would rather use what they know than be forced to learn what another app... even if it means that the fragmentation of for example structural analytical models. Convincing your BIM manager guy is not going to help here as he will not be able to argue against the expensive down time that cross training would entail.
The other interoperability problem is longitudinal in nature. I see a lot of engineers who have invested a lot of time using 'generic' computational design tools like Grasshopper to design things like bridges, and get patted on the back for this, as this is seen as innovative. The problem is how to productively hand over the preliminary model downstream to other apps for more detailed design and analysis. When asked why do they use GH when there are other specialist apps like OpenBridge Modeler etc, they say they have never heard of OBM... or it's too complex... and even after acknowledging that at some point the heavy lifting needed for demonstrating structural code compliance would overwhelm their 'home made' model. After suggesting that the innovative engineer should ideally collaborate with Bentley to build in the algo's on top of the base vertical (ie let Bentley worry about debugging, maintenance, Windows compatibility etc) , the comment / rationalisation is that they do not want their IP ending up with Bentley.
From what I have seen, Bentley professional services in London are failing miserably. Seems like you have to be a major engineering outfit in an emerging market like ECIDI in China before you will get any real support from the mothership. It seems that here in the mature markets, all you get to do is fill out a Service Request, and get in queue, waiting for Godot.
Frankly, you shouldn't be making this pitch. Bentley should be, backed up with some dev muscle, not some sales / support rep, who are just post boxes that do not understand or sit with the engineers or code. Get in there and embed full time support staff and have regular meetings with the dev team in India etc.
Jon Summers said:How do I counter arguments along the lines of 'Revit already does that'?
Yea... Revit has a lot to offer. It has something similar to EC with its Extensible Storage API and SchemaBuilder (equivalent to Bentley ClassEditor?). Revit also has long standing API's that cover parametric change propagation, and dynamic updating that allows third party apps to participate in the dynamic updating.There does not seem to be any comparable overarching framework in the Bentley camp. Every app dev seems to be responsible for the way the app behaves in isolation; and this means third party apps will not be able to interact with each other.
Sure, I suppose that you should be able to argue that Bentley has been at this a lot longer with ECM, ISM and ISO15926 etc, but as mentioned above most engineering firms are not very savvy, digitally. And they are still wow'd by the data-centric way Revit can propagate change (I guess that they see this as labour saving and cool automation) and how dumb and old skool CAD-like Mstn's behaves in comparison.
I think that you will need to consider where the firms that you are speaking to are on the 'digital' ladder. Top firms like Bechtel or ECIDI or China Power etc already have substantial investment in coding, and have full blown home made apps. Bechtel has iCOMET for plant design. ECIDI, China Power etc have some pretty impressive in-house apps as well if the Be Inspired presentations are to be believed. Probably in markets that can justify ample / expensive input from the primary vendors or re-seller channel or somebody like GeometryGym.
Other firms like Arup even have full blown commercial apps like OASYS that have been spun off. Thornton Thomassetti has its TTX suite which is aimed at interoperability tasks. These days, I think things like Flux and FME are probably the go-to guys for developing interoperability tools. I suppose that it shouldn't be too hard to provide an interface with these platforms.
Next step down: are firms that do not have full-time coding arms, and have a lot of little VBA's and other add hoc automation tools using Excel or a database to translate / map data between apps, with the help of a lot of interns. There should be a lot of scope for Bentley to offer some help here.
Jon Summers said:Where can someone find examples of core technology EC Schemas emerging in vertical apps. such as AECOSim?
I think the best place to start is looking at the OpenPlant forum. Not Aecosim. Aecosim suffers here due to fact it is almost only used in the infrastructure sector, which lags way behind the both plant/process/petroleum and even the building sector as far as BIM and schema information is concerned.
Roads and rail: IFC5 is still in development. I understand mainly by the Chinese, Koreans and Germans.
Water: Not sure what the status is, but the emphasis seems to be creating component catalogs, with in-built manufacturer's data. There has been a couple of examples of firms doing this presented at Be Inspired over the years. Forced to use OpenPlant due to the need to provide P+ID information... but I know that one firm was using Revit to model the components because the CAD guys did not know Mstn, and Aecosim's PCS was so impenetrable and on hold awaiting replacement. Weird decision since it will be more likely to find dwg or dgn based apps used in plant and drainage design... so, all that Revit-based content will need to be converted to be used. What a way to built in interop overheads and a typical example of the lack of forward planning by top management !
One assumes that these water treatment plants would be based on an analytical or process model package like EFOR, BioWIn, ASIM, AQUASIM, SIMBA, WEST etc (?), so the schemas will have to follow the analytical modeling package's requirements. Bentley's AXSYS was not mentioned at Be Inspired, so I assume that it hasn't been used in the 2-3 case studies presented.
From what I have seen, the analytical models are not connected to the BIM model in the AEC sectors... even with big engineering firms. Appalling to see huge public health models being built in Aecosim that are so completely devoid of any analytical smarts... that they have to manually checked by a guy using a spreadsheet and calcs on pencil and paper!! Big opportunity for Bentley Pro Services to step in and help automate things.
GIS: I think that Landxml and Esri were the early vehicles for this. Most infra jobs have extensive GIS or regional modeling requirements. Things like transportation capacity/congestion, land use / economics, environmental / pollution impact, noise, acoustics, energy consumption all have to modeled and evaluated , which will need schemas. I doubt any of this has been normed under an industry body. And the modeling packages are probably few and far between, so each project will be more likely to just plonk for one package for each task, and rely on ad hoc translators and platforms like FME for the tricky bits that need niche 'foreign' packages.
Electrical: Not sure, but my impression is that the focus at the moment is defining schemas for facilities maintenance, eg IADD4UK, if anything. This is more about asset tagging than about schemas aimed at interoperability. AEC3 Nick Nesbit et al has been at this for awhile. Maybe you should team up.
How would an ISD pitch something like EC schemas? Sure, schemas will make interop easier, and if you could offer a turn key solution... then it would just be the usual haggling. But, if you can not offer one and it is more like offering to 'co-create' a solution for a fee... it is much harder to define a viable business case. The starting point is usually a client requirement. At the moment this is a pretty hairy proposition as most client bodies are pretty ill-informed about what they want. Some would throw up their hands and say that they are incapable of defining their requirements and hide behind all kinds of woolly, open-ended and aspirational clauses in their contracts; and continually come back demanding more every time they hit a snag. Maybe you should speak to your BIM-M friend and go through the bids he is involved in to see how you can help formulate the RFP response to manage expectations; budget for it etc.
And this is just the deliverables format, never mind the actual tool / user interface and data management workflows that would need to follow on and shape the designer's tools... if we are to avoid drowning in the additional work required.
I would say that Bentley are pretty hopeless with this kind of thing, where lots of hand-holding on site is required. The re-seller channel has been relegated to the 'left-over' small or medium-sized firms, making them least exposed to and capable of dealing with large tasks like this. And it also appears that the ISD 'ecosystem' has been progressively shut out and starved of access to the inner workings of the Bentley platform... leaving a vacuum for others to fill. Bentley has probably shot off a foot going ten years back and have been hobbling off into the sunset without realising this, ever since. For an ISD, this will mean investing the time to understand what Bentley can already provide. No proint re-inventing versioning if ISM already does this, or Component Indexing, full text search etc etc No point re-inventing KBE tools like Design++ if the tools are already there. Maybe you should be on retainer to a local Bentley office supporting the sales reps and demo jockeys?
As mentioned, there seems to some bold thrusts in China, with Siemens and others (TopCon?) probably personally sign-offed by one of the B-Brothers, where schemas will play a role... but can this be leveraged more generally?
Interesting piece by Bentley colleagues from China on how EC relates to Bentley apps.
•ECObjects and ECXML: used to define engineering content (for representing EngineeringContent); I suspect that probably easier to re-write OBD functionality to Mstn (See Linear Solids, Parametric Cells, Constraints, Named Presentations, Perforators, Dynamic Views, Element Templates etc) than the other way round?
•ECServices: It is a set of unified services provided for the engineering environment of multiple data sources; Some usage for interop, Edit in Excel?
•ECControls: EngineeringContentBrowser, PropertyPane and other controls; Some overlap for dialogs?
•ECClassEditor: an editor for externally editing ECSchema; Replacing DGS by something closer to what OpenPlant uses?
ECRepositories: I wonder if iTwin is next as part of "Phase 2"