28 July 2015

Procuring BIM - PAS 1192-2 and acif PTI

I feel sorry for owners and managers who need to make decisions about BIM on a project.
The information available is vague; it is hard to extract practical advice that can be acted upon. And confusing, mixing up BIM issues with management practices that have nothing to do with BIM.

PAS 1192-2

I recently worked my way through the UK PAS 1192-2:2013. Full title: Specification for information management for the capital/delivery phase of construction projects using building information modelling.
It is a "Publicly Available Specification", which are documents created for a sponsor by the British Standards Institution (BSI). In this case the sponsor was the UK Construction Industry Council (CIC). Not that the BSI do all the work. The PAS is done via consultation and a number organizations have been involved (24 are acknowledged). It was published in February 2013 and is 68 pages.

Its audience "includes businesses and those responsible for the procurement, design, construction, delivery, operation and maintenance of buildings and infrastructure assets."

What interested me is that PAS-1192-2 is an attempt to holistically capture BIM processes from beginning to end of a construction project. It is one of the few examples which tries to proscribe how to commence a BIM project, how to create a BIM brief.


acif BIM & PTI

The other documents I recently slogged through were a series on BIM by the Australasian Construction Industry Forum (acif - they prefer lowercase), a peak body of peak bodies, including the likes of The Property Council, Engineers Australia, Master Builders, Facilities Management Association to name a few. The BIM documents are authored by the Strategic Forum for the Australasian Building and Construction Industry, a body within the acif that that "brings together key stakeholders".

Documents include:
A Framework for the Adoption of Project Team Integration & BIM, published December 2014, acknowledges 11 participants, and is 60 pages.

Building and Construction Procurement Guide: Project Team Integration and Building Information Modelling (BIM), published June 2015, acknowledges 8 participants, 56 pages.

These documents essentially cover the same ground, with the later one containing marginally more specific 'advice'.
The first "is designed to guide and assist industry stakeholders in the adoption and implementation of PTI and BIM."
The later "is to provide asset owners and project procurers with an outline of potential procurement practices, processes and steps which might be followed in developing effective procurement strategies for implementation of Building Information Modelling (BIM) and Project Team Integration (PTI) on specific projects within the built environment."

There are also two documents on Project Team Integration (PTI): The Case for Project Team Integration and Project Team Integration Workbook

Despite their titles and self descriptions all of the acif documents are more BIM and IPT sales pitches than practical advice or structured workflows.




It is interesting to look at these documents side by side as the UK is heading for mandatory BIM, whereas Australia is, well, on its own. The current federal government doesn't believe anything needs to be done about global warming, so BIM is way too avant-garde for them to even comprehend, let alone mandate.
But governments change, and what the acif is spruiking may end up in the form that PAS-1192-2 takes, or indeed the wholesale implementation of an unchanged PAS 1192-2.

So how do they stack up?

AN EASY READ?

In a word, no.

PAS 1192-2 is acronym city. I had to spend a lot of time memorizing the myriad of abbreviations:
BEP, TIDP,  MIDP, RM, PlM, PIP, SCCS, SMP, CPix, EIR, Capex, Opex, CDE gates, RACI, WIP, AIM, CDM,  and not only LOD but also LOI.

Although PAS 1192-2 has a glossary, not every acronym is listed. It wasn't until page 13 that I found out what PAS stood for.

Terminology varies to a frustrating extent. You kind of expect some variances, particularly as PAS 1192-2 is from the UK, acif documents from Australia. But the authors seem to revel in creating their own unique terms. The Owner is called "Employer" in PAS 1192-2, "Project Sponsor" in acif documents. acif have invented a new term for Integrated Project Delivery (IPD) - "Project Team Integration" (PTI). They explain that PTI is a more generic term, IPD being a form of contract rather than a description. I don't see it. IPD is already a term in use in Australia, and perfectly adequate. Why invent a new one?

PAS 1192-2 is a document you study - take notes, reread sections, draw your own diagrams, google a lot (to find out what the acronyms mean). Just reading it will leave you confused and be of no practical benefit. A lot of thought has gone in to it, but golly, does BIM have to be this complicated?

The acif documents are easier to read - if you can stay awake. The same things are constantly repeated, not just within documents, also across documents. There is really no point reading the Framework document, the same information is repeated in the Building and Construction Procurement Guide.

But to be fair all documents of these types are tedious to read. Both sets of documents are extremely well structured, have good contents pages, glossaries and definitions. And unfortunately repeating information is de rigueur for these types of reference documents.

REALISTIC WORKFLOWS, OBJECTIVES?

PAS 1192-2 describes workflows, the acif documents really describes objectives, and now and again, if you know what to look for, actions to achieve objectives.
Generally both have a good grasp of BIM, with inklings of evidence there are at least some people involved with direct experience. But there are some areas that I believe push the bounds of practicality.

PRE-TENDER BIM PLAN

The requirement for a pre-tender BIM plan in PAS 1192-2 is unrealistic. I don't see how this is even possible unless the tender process is severely limited to only seeking bids from consortiums. A BIM plan needs all parties to get together to agree on a plan. How can this be done at tender or RFT stage when multiple parties are bidding for the same work? If it is enforced it sets up an environment where collusion could run rife. All those tenders getting together and just discussing BIM?
Unless, that is, PAS 1192-2 really means something other than a traditional BIM plan. I searched for this possibility but could find nothing that suggested otherwise.

But I smell a rat. Although PAS 1192-2 explicitly states it is suitable for all contract types the underlying assumption that comes through is that only IPD (Integrated Project Delivery), Alliance and other combined risk contracts are suitable for BIM.

This assumption is also explicit in the acif documents. Their introduction of a new acronym - IPT, (Integrated Project Team), and the time spent extolling its virtues belie their underlying intent. More on that below.  

GATEWAYS OR BLOCKAGES

PAS 1192-2 includes "Gateways", where BIM data is approved before the data is released for use by others. This is not a new invention brought about by BIM, many QA processes already contain such procedures.
Hold and review points are good in theory, but if not structured and managed carefully can end up being blockage points instead. In my experience the reality does not always match the intent:
  • Adding review points should extend the program, but this never seems to happen (what owner volunteers to extend the completion date for the sake of a technology?). Instead work programs are condensed to unrealistic levels to the point that work continues into the review period, leading to poor work and inadequate checking.
  • Owners, or more usually the poor sods they appoint to oversee all this checking, don't want to take any additional responsibility (nor workload) so refuse to, or drag their feet in officially signing off on anything. So work continues on, as it has to, using unapproved deliverables.
  • Your boss sees checking as a non-productive use of time, so if some-one else is doing it why are you doing it? Leading to unchecked documents leaving the office. 
  • Any review point is an opportunity for designers to "tweak" the design, leading to rushed reworking. 
These problems can be addressed, but PAS 1192 seems ambitious when it comes to sign off at Gateways (which it admits may be difficult for "some contracts").
The problem in PAS 1192-2 is sign off is assumed to mean a shift of responsibility from author to approver. Any mistakes become the fault of the checker for not picking them up. Which means the checker must have expertise in the area they are checking. So an owner needs to appoint a second architect to check the primary architect's work, structural engineer, services engineer etc. Not very efficient.

There is a trade off between mistake free documents and project progression. To ensure comprehensive and mistake free documents takes considerable time at each check point.
A more practical approach would be to check only for completeness, the professional risk still being carried by the author.  

VOLUMES

PAS 1192-2 has a concept it calls "Volumes". The idea is that the project is broken up in to a number of volumes (3D spaces) that are allocated to different project team members. The example of a rail tunnel is illustrated with linear volumes for different services.


This may work for simple infrastructure but I don't see how it works on a even moderately complex building. Different services often share the same space (e.g. ceiling space). Allocating specific space for different purposes is possible, but is generally not the most efficient way to design a building.

Further "all members of the design team shall agree volumes as fully as possible at the start of the project". How can you do this before designing the building? The allocation of space is a huge part of design, most of what an architect does. Does PAS 1192-2 assume the architect's work is complete before BIM is started?
It wouldn't surprise me. There is a pervasive belief that BIM only starts once a contractor gets involved. Part of the absurd push for IPD (aka PTI): that BIM is only possible if the contractor is involved during design.

How about the acif documents?
These documents are mostly 'motherhood' statements with a sprinkling of useful advice.
For example on page 31 of the Framework document out of 15 objectives;
  • 3 are general statements ("the dismantling of traditional barriers or silos of effort")
  • 3 are not relevant to a project but to the industry as a whole ("further development of national  templates, content and Standards")
  • 2 are repeats of issues already stated.
So just over half are not useful. One wonders why they didn't have two lists of objectives, one for industry and one for practitioners.

Recommendations made years ago reappear, like "undertaking pilot projects to display the benefits of BIM." Not more pilot projects! How many more BIM events must we sit through where all you get are syrupy presentations of (apparently) extraordinarily successful BIM projects.

And there are contradictions. Under Asset Management "Proposed Activities to deliver on Objective" one 'activity' suggests another is not possible.
How can:
"A contractual obligation (clause) binding on all parties from initiation of a built project for the development, transfer and maintenance of an asset register across the asset life cycle."
be achieved until:
"The asset/facilities management industry must define data sets and information asset register outcome requirements to enable the transition from design and construction to operation in a BIM environment."
It is a classic chicken and egg situation. How can AEC team members provide something that is undefined?

THE COST

There is quite a bit of extra work for owners ("Employers") in PAS 1192-2. From being responsible for proscribing the entire BIM process to checking it has been complied with. To follow PAS 1192-2 owners will have to beef up their project management teams, not just in training and expertise, but in bodies on the ground to do the additional work.

PAS 1192 also introduces a raft of extra requirements for tenders. Although the owner may not directly pay each tenderers for the additional work, the industry as a whole will need to recoup those added costs.

COBie deliverables and assignation of Uniclass classification codes are mandatory, even though there may be no-one using these on the project. Why provide COBie if an FM solution is part of the construction contract and data can be placed directly into the chosen FM system? Sure, preference COBie and Uniclass coding where FM data and cost coding are required, but only if team members have no viable alternative.
As I have written in earlier posts, both of these imposts create additional work. Work that needs to be paid for, whether directly paid for by the owner or as a cost to the industry as a whole.

Sure BIM may bring savings elsewhere, but strict compliance to PAS 1992-2 will be an additional cost. Therefore be wary of statements like "must comply with PAS 1192-2". Owners making statements like this are adding possibly unnecessary costs to their projects, others with it in their contracts need to make sure they have allowed for the extra work in their bids.

The acif documents are not proscriptive enough to identify where there might be additional costs. As they are primarily about introducing BIM the most obvious cost is in education and training. Although the implicit assumption throughout their documents that owners must take a bigger role in BIM is a potential additional cost for owners.

DOES BIM DRIVE THE FORM OF CONTRACT?

Building contracts are structured to achieve many outcomes, and attempt to create agreement on many issues, BIM is only one, and is by no means the most important. Yet both PAS 1192-2 and the acif documents assume that BIM processes can only be achieved under one contractual arrangement.

Interestingly both PAS 1192-2 and the acif documents specifically state that they are contract neutral, acif documents even warning that owners ("Project Sponsors"):
"... need to be careful that changing contractual arrangements for BIM doesn't lead to a degradation of other aspects - like design, innovative construction, innovative engineering solutions."
But when you read the documents as a whole it becomes obvious the only way the requirements (PAS 1192-2) and objectives (acif documents) can be met is with an IPD type contract.

In the acif framework document there is a good description of different contract types, including existing "traditional" contracts. Then there is a table comparing contract types with their effect on BIM Implementation. Except they lumped all existing contract types together and compared them to alliancing ("partnering") and consortium ("financing") contracts. What would have been far more interesting, and actually useful, would be to compare BIM implementation between each of the existing contract types (Construct, D&C, Managed Contract, Construction Management etc.).

l don't understand where this idea comes from that only certain types of contracts are suitable for BIM.  Any contract type can use BIM. The truth is (as mentioned, but contradicted elsewhere in acif documents) the form of BIM is set by the type of contract, not the other way round.

BIM AS SOCIAL ENGINEERING

As mentioned PTI (Project Team Integration) is a substantial part of the acif documents. There is talk of  PTI Protocols but I couldn't find anywhere that lists or describes what these protocols are. There is a list of their purpose, and why they are important but not what they are. Are they talking about specific existing protocols, future protocols, a protocol, or a series of protocols?

I got excited when I found the acif Project Team Integration Workbook. A workbook, something practical, something that should tell me what PTI is.
Sadly I was mistaken. It is a series of 18 tables of generic project management topics, like "Environment and Culture", "Project Leadership", "Wasted Effort". Each is divided into 5 colour coded columns, red is bad, blue is exemplary.


You guessed it, existing contract types only appear in the red column, PTI type contracts dominates the blue. Amazing, doing PTI (whatever that entails) will miraculously make everyone a better manager!

Choose PTI and your project goes

from    "This is the worst project I've ever worked on in 30 years"
to        "This is the best project I've ever worked on."

from   "We're at war. The client's the enemy"
to       "We have the greatest respect and admiration for our client. He leads without interfering."

Hallelujah, praise to the god of BIM. All management sins will be washed away by accepting the wisdom of PTI.

WHAT IS THIS PTI?

In the acif Framework document section headed "Agreed definition of PTI and BIM" it states:

"PTI is a process to facilitate integration and encourage collaborative behavior ..."

But what is this process? It also states:
"PTI is a project delivery approach that encourages clients to engage a team (including design consultants and building contractors) at the earliest stages of a project to enhance the level  of integration between them."
OK, the team gets together early. But what explicitly do they do that is different, to make it PTI instead of business as usual? Besides more motherhood statements like "reduce waste" and "optimise project outcomes" there is nothing about what specific procedures constitutes a "process".

The give away is the word "collaborative". This is nothing more than another version of the "we must collaborate" myth I have written about in other posts. I do not know, and have been searching for, what I should be doing beyond what I, and the people I work with, already do to achieve this "collaboration". The only logical thing I can get a firm grip on is the idea that we should be providing additional information for others to use, which I call for what it is, exploitation, not collaboration.

But I don't believe that is what is behind the acif documents. I think they are under the impression they can foster a revolution in the quality of construction project management through the adoption of BIM. My suspicions were reinforced when I read the acknowledgements in the acif PTI Workbook document. There are no practitioners of BIM mentioned, and it is based on a 2001 publication by two academics:  "Projects as Wealth Creators".

I'm sure their ideas for better project management are fantastic and worth adopting, but they are trying to hijack BIM to push for unrelated issues. Using a new technology to justify a call for social change, otherwise known as social engineering.

BIM is not going to change the way people behave. An owner who tries to squeeze everyone's prices and goad them into extra work is still going to do that. This is already happening with BIM in the attempts by owners to get the AEC team to provide FM data in a format of their choosing at no additional cost.

And it simply won't work. No manager thinks what they do is poor practice. When they read something like BIM requires "a well informed client who knows what they want" they never think it applies to them. Statements like that achieve nothing.

What should be explained is that anyone can utilize BIM, it is just that better managers will reap better rewards. That those who are open to adjusting their management style will benefit more than those that don't. To say to a manager you must throw out how you have done things in the past to use BIM is not only untrue, but counterproductive.

SUMMARY

To summarize my main criticisms:

PAS 1192-2

EMPLOYER (Owner)
-Assumes all is known about the project before it starts.
-Expects employer to define how professionals conduct their business.
-Expects employer to have expertise to check and approve professionals work.

PROCESS
-Assumes design team all engaged at same time.
-Assumes RFT from consortium - how else does a "pre-contract BEP" get done?
-Assumes IPD type contracts (even though says it doesn't).
-Assumes building has been designed (e.g. expectation of "volume" definitions)

DELIVERABLES
-Insists Uniclass codes allocated to BIM objects even if not used by the project team.
-Insists on COBie even when there may be no reason to use COBie

acif DOCUMENTS

-Doesn't explore how BIM can be used on existing contracts.
-Mixes project specific objectives with industry wide objectives.
-Many objectives are not actionable.
-No decisive description of PTI protocols.
-Inconsistent: advice is often contradicted by other parts of the document.

CONCLUSION

You may, by now, have noticed this post is not a precis of the documents reviewed, but a critical review that carefully avoids spoilers (sorry, you are going to have to read them yourself).

But having read them I believe both PAS 1192-2 and the acif documents are worthwhile additions to BIM literature.

My criticisms are born of a frustration, that they are so close but miss the mark.
My belief is that there is enough knowledge out there to create useful, practical BIM guides. The reasons why this rarely happens lie elsewhere. You can literally see the tussle between the BIM practitioners and BIM evangelists as you read the acif BIM & PTI documents.
A good edit cleaning out the myths and rearranging the truths would do wonders for these documents.
I believe all PAS 1192-2 needs some time in the real world getting practical experience to see what works and what doesn't. Some parts probably need to be watered down but the underlying logic and workflows are sound.

So my advice is to make use of PAS 1192-2; for guidance on setting up a BIM procurement process, and the acif documents; for an overview of possible BIM procurement methods, with some gems of  practical advice.

Just don't ever mandate that PAS 1192-2 'shall' be followed, and approach the acif documents with a dose of skepticism, mine it for the gems and leave the tailings behind.