Decomposing Services

I had a very nice conversation with Bob Smith of Tall Trees yesterday about building services. Bob is one of my co-conspirators launching the Building Service Performance ONTOLOG group. Bob had just submitted the laundry list of check-offs developed by the City of Irvine for its construction process to the ONTOLOG discussion.

I was both happy and disturbed to receive this document. These check-offs clearly drive the contracting process, They are also inherently backward looking, enshrining the best practices of yesterday as we move forward. The best practices of yesterday are better than the normal practices of today, but can be the enemy of better practices going forward. They trade innovation for good enough.

The ONTOLOG (just google it) project is to define an ontology of Building Service Performance. The problem we are trying to resolve is that while people want specific services from their buildings, we always specify technologies or systems, which is something quite different. Buildings may be providing alert students, productive office workers, or regulated environments to store labile chemicals. By discussing the services rather than the systems, we can

• Allow earlier discussion of / decisions on building goals in keeping with buildingSmart approaches
• Move conversations about building performance to the business level where commercial building owners will get interested (we want to provide healthy office space metrics in the upper quartile while staying within energy use goals).
• Commissioning then gets re-defined in terms of service performance effectiveness rather than in terms of system operations, a much more useful measurement.
• Commissioning numbers that look like B are simple enough to get built in to the sales and leasing processes for commercial buildings, enabling owners to monetize improved performance.

Check off lists such as Bob provided are the opposite of this approach. Nonetheless, I agreed to try to dig up two other documents that are contrary to the goals and thinking of the new group.

One such document is the original list of nearly 50 vertical control system markets that we came up with as oBIX was launching. These systems are specified in buildings now, but only rarely with any sort performance or business deliverable tied to them.

The other document I am trying to come up with is a comprehensive list of ICC (International Codes Council) domains. It struck us during conversation that many of the ICC areas deal with avoiding the failure to perform these services. While we are trying to twist these services around into a new ontology, a list of all the services which must not fail to be provided would be useful.

I have found neither list yet. If you think you have one, please send it along.

Why BIM, Why Now

When I prepared my talks for REALCOMM this week, I had assumed that many of the attendees, active in REITS and large property managers, would be familiar with Building Information Models (BIM), and in particular the Nation BIM Standard (NBIMS) and buildingSmart. Regular readers are familiar with these, but because some new readers may be coming from the conference, I am going to review quickly why the property owner should know BIM, and demand buildingSmart.

The traditional business process is designed around delivery of hand drawn paper. CAD automates paper production, but not the completion of business process nor the tracking of deliverables. CAD only works for simple buildings, without fine tolerances and building controls. Because bits of information are spread over multiple two dimensional (2D) drawings, issues are not properly resolved in advance. Duct may be routed through a steal beam. Plumbing may block the telecommunications chase. Multiple utilities may not even fit above the dropped ceiling.

Capital projects need to adopt the disciplines and use of information technology adopted by manufacturing in the 80’s. It takes diligence, experience, and time, to catch these problems in a CAD project; in a large complex project, the task is too difficult. Usually they are caught during the construction process, creating additional expense in the form of change orders and construction delays. BIM is the answer.

Best Practices in Information Exchange relating to Capital Projects are codified in the National Building Information Model Standard (NBIMS). NBIMS aims to create a common standard for electronic exchange of all information associated with a building new or old to be able to use that information in all stages of the facility life-cycle. This information model allows the free flow of graphic and non-graphic information among all parties to the process of creating and sustaining the built environment.

Newly valued deliverables are not linked to traditional business process. For example, Energy Modeling, has no intrinsic link to design. The designer is not liable for it. Using BIM, energy models can be generated directly from the design, in effect commissioning the design before it goes to bid.

The following are symptoms of the poor information exchange using traditional design approaches:

• Failure to adequately integrated complex systems during construction process
• Late construction change orders
• Failure to hit Energy Modeling targets
• Difficult or undefined Commissioning targets
• Incomplete designs, especially of control systems, mean that there are many modifications during construction; these modification increase the cost of obtaining accurate as-built information.
• The many changes during construction make it difficult and expensivce to get as-built drawings of completed projects.
• Ill-defined and ineffective hand over of operating instructions mean that information is inaccessible.

NBIMS produces the following immediate benefits:

• Standardizes construction documents in line with national and international standards
• Models rather than draws buildings so problems are caught before construction, when they are cheap to ameliorate.
• Run Energy Models directly against Building Models to enable iterative energy analysis, even during value engineering
• Contractors are able to bid electronic Building Models, not the piles of paper.

ASHRAE has published the following numbers on the benefits of using a building model for construction.

Design:

• 20% to 50% reduction in Design Cycle Time
• 100% Accurate Procurement package

Construction:

• Time and Cost reductions 20% to 40%
• Reduced Rework
• Elimination of most change orders
  

Operation:

• Life-Cycle O&M reduction 10%-40%
• Reduced Handover/Turnover time

I don't see how you can afford to not use BIM on your next project. Please write me if you want to know more.

Open Source oBIX advances

From: Peter Michalek <petermic@gmail.com>
Date: Mon, Jun 9, 2008 at 10:12 PM
Subject: oBIX server 0.2.04 released
To: obix-developers@googlegroups.com


Hello everyone,

Version 0.2.04 of oX oBIX server has been released.

It adds experimental support for external modules, a stand-alone sdk that simplifies module development and expands the documentation.

The feedback to the original specification asked for the following feature:

"My big wish, make it possible for oBIX clients to add data to the server so clients can integrate with each other using this server"

Providing an experimental implementation of external modules is a step to achieve that. Although the sample in ox-sdk uses java, you could use any other language such as python or C to accomplish that, since the interface (which will evolve), is REST-based PUT.

There is a quick start that makes it easy to try the server out:
http://obix-server.sourceforge.net/oxUserGuide.html#chapter-quick-start

There is a step-by-step set of instructions of how to build your external module.
http://obix-server.sourceforge.net/oxUserGuide.html#section-Module-Creation-Walk-Through-External-Module
http://obix-server.sourceforge.net/oxUserGuide.html#section-Module-Creation-Walk-Through-Internal-Module

To read more, please go to http://obix-server.sourceforge.net/ which also contains pointers to downloads and release notes.

I'll keep updating the demo server with each release: it will be running two simulator modules:
http://obix.michalek.org:1225/obix/
http://obix.michalek.org:1225/obix/about/productVersion/

obix client:
http://obix.michalek.org:1225/browse/

Your feedback is greatly appreciated.


Peter


--~--~---------~--~----~------

Biodiesel Algae for the Building

I was corresponding with someone from the algal biodiesel group the other day. Genetically modified algae is one of the more intriguing fuel strategies in the mid-term. The short version is to add some oil-production genes from some other plant to fast-growing algae, scoop out algal mats and process into fuel.

Traditionally, algae has been seen as something to grow in plants about the size and distribution of this year’s boondoggle, the corn ethanol plant. Instead of large parking areas for constant transportation of corn, large shallow vats of algae would soak up the sun. Eliminating the need to transport the raw material to the processing plant would be yet another advantage to this process.

Some have suggested that the proper place to build the facility is by a coal plant. Algae grows faster in a high CO2 environment. The CO2 would get sequestered into new biomass, and then converted to biodiesel. The CO2 would make it into the atmosphere eventually, but not until it had done double duty for electricity and transportation.

But I thought, why stop there?

All kinds of moderately complex processes are now being built into small microprocessor controlled autonomous systems. If one could automate the production of Biodiesel on the rooftop, then local diesel generators could run on site generated fuel.

I do not imagine that this process would ever provide all power for, say, a commercial office building. It could, however have a place in zero net energy buildings and in local self-reliant microgrids.

Many organizations, from the AIA to ASHRAE, from the US department of Energy to the UN Environmental program, are chasing after the Zero Net Energy Building (ZEB). The ZEB uses a variety of strategies centering around local generation, storage, and conversion of energy to limit its purchases from the power grid to when the prices are right. The ZEB will likely make use of internal DC to eliminate DA/AC/DC conversion penalties on each source of energy. The ZEB building may well have PV, ST, Wind, and generators, mixing and matching as needed.

The problem with most of these local renewable energy sources is that they are unpredictable. As has been well demonstrated by the German Kombikraftwerk effort (search the archives), you can build a reliable grid almost entirely of unreliable sources as long as they are unreliable in different ways at different times.

Why not BioDiesel generators in the building? Why not algae vats and automated fuel production in the building? I do not see such a system being able to carry the building on its own, but if called on occasionally, as diesel generators are now, perhaps the tank could be filled in the interval.

So, why not Algal Biodiesel in the Building?