One of our goals at Beck Technology last year was to conduct a second comparative analysis of design-build proposals using DESTINI Profiler. The first time we used DESTINI Profiler to compare contractor proposals, we were asked on short notice by the US Dept. of State to review responses on a MONSTER-sized project. The job was valued at close to $1 billion and included 25+ existing buildings to be demolished and then replaced in a highly orchestrated fashion in order to maintain operations of a US embassy. On this project, the Beck Technology team members had 24 hours to take an existing DESTINI Profiler master plan model that they had created during planning and integrate it with the proposed schedules from 2 competing contractors plus the government’s own internal phasing plan. The results of our analysis were pretty amazing (see video here), but we wanted to see if this process could create value on a smaller, less complicated project.
The opportunity finally came when I met Bruce Rainey with Scripps Health at the COAA conference. I approached him with the idea and several months later he identified a specific project and connected me with his project manager. The project is a 175,000 SF outpatient facility in La Jolla, California. It’s to be connected to an existing hospital on two levels and has six above grade levels and one level below grade. The building would require changes to the roads and surface parking around the hospital complex and this work would need to be done in a phased manner in order to minimize disruption. Two existing buildings also need to be demolished to make room for the new clinic and are a part of the proposal. The project budget is $68 million.
Prior to receiving the proposals, we created a basic DESTINI Profiler model of the project based on the same documents that were available to the contractors. Our model included the new building with floors, cladding, and site paving (existing was to be demolished). Interiors were not modeled. We also took the owner’s budget and schedule that had been developed by their cost consultant and program manager and applied it to the model components. We did this using DESTINI Profiler’s Excel import tools which we configured to use Scripps’ standard cost categories. All the prep work was completed in one afternoon setting the stage for the comparison; now we just needed the proposal data.
Our meeting with the client to review our results was scheduled on the Monday following the Thanksgiving break. We had planned to get all the information the week prior and have plenty of time to put together our assessment and the collateral to present our findings. That time frame worked out well for us. Unfortunately, the bid information didn’t make it into our hands until the day before the break. Consequently, the work we did was mostly done while I was at my in-laws watching Thanksgiving football games (luckily Texas A&M put Mizzou away before halftime buying me a couple of extra hours!! Whoop!). The total number of man-hours invested on the DESTINI Profiler comparison was approximately 24.
The steps were pretty straightforward on the costing. First, we had to create a version of the model with each contractor’s cost information. This process was made easier because of the spreadsheet that we had created earlier. We simply entered values in new columns in the original Excel file, imported the new values, and “saved as” each version of the model. This process also gave us a nifty comparison in Excel where we used color to identify extreme cost differences.
The schedule info was a bigger challenge because the level of detail, description of the phases and number of tasks varied widely. This process was also done in Excel utilizing the sequence import function in DESTINI Profiler . Tasks were summarized from each submitted schedule and similar milestones were identified. Each schedule was boiled down and listed out in our format then applied to the model. Some minor tweaks to the model had to be made from one plan to the next based on the sequence of the tasks to account for differences in how the work would be put in place.
At this point, we had four unique cost and schedule loaded DESTINI Profiler models (including the owner’s plan). The next thing we needed to do was to create a 4-way side-by-side comparison and then “play” it to see the differences. To do this, each model’s data was exported and then combined in PowerPoint using an automated combiner tool built by Beck Technology. The export created a 3D image at each time increment (in this case monthly), showing active tasks, cash spent, and a couple of cash flow charts.
The resulting visual analysis was just a matter of hitting play and viewing the outcome. When watching the results for the first time, we immediately noticed that one of the contractors jumped out to a very quick start relative to the others. This raised some flags regarding the known permitting bottlenecks they normally expect. It raised the question, “Was this realistic?” We then observed that one of the contractors had scheduled a phase of the roadwork starting before the building demolition had been completed … in the same space. This was a physical impossibility, but as obvious as it sounds, it was not obvious on the traditional deliverable. The owner received a 20-page Microsoft Project PDF where the road work was on page 20 while the demo work was on page 2 which was undetectable at a glance. It is almost a guaranteed change-order! The final major difference we observed had to do with the sequencing of the exterior façade. All the contractors except one sequenced the exterior cladding from the ground to the roof, floor by floor. The outlying contractor saved some time by installing the cladding one face at a time. This reduced the number of crane moves and thus reduced the overall schedule; however, the concurrent interior construction created a concern. The model showed that the interior finishes were well underway while the building was far from being dried in. When we pointed this out to the owner they said that this activity was proposed at the peak of their rainy season and it would almost certainly cause weather damage.
The most significant issue with the proposals was that they all came significantly over budget. Proposals ranged from $78-83 million while the budget was $68 million and the owner had no ability to go get more funding at this point. Beck Technology’s goal is to be involved earlier with DESTINI Profiler next time to help them set a more realistic budget to begin with. Then, we want to use DESTINI Profiler in a Target Value Design process to help them keep their concept on that budget! At this point, the winning contractor would have significant “value engineering” to do upon starting their work to reduce the cost by approximately 20%.
Our analysis was presented to the program manager, project manager, and other owner stakeholders. The feedback was that it improved their understanding of the proposals and increased transparency in the review to a higher degree than they had ever done before. As a result, the owner made a best value choice that so happened to be awarding the project to the slowest, most expensive bid. This contractor’s proposal had the fewest holes and was the most conservative from a risk stand point. It was comprehensive where others listed exclusions for various items (such as toilet paper – not a joke!). In the end, we were excited to have been a part of it. It gave us some real insight into the review process and mindset of the owner and project stakeholders. It also validated our technology could be used in this way on common commercial projects – not just complicated mega projects.