A fully-connected construction industry is long overdue, but many technologies and integrations still need to be made that are long overdue.
A country’s construction industry can account for 6-9% of their total GDP. However, across the board, it has shown weak or negative productivity growth compared to other industries over a thirty-year period. This is due to a number of factors in the United States in particular, including:
- Older equipment, availability of consistent construction drawings, and lack of direction from on-site staff.
- The use of unskilled labor, poor managerial techniques, and lack of available material.
- Lower profit margins for construction companies in general.
- Poor site and working conditions in general
There have been a number of changes to make the industry more nimble, including the application of BIM. For more complex projects, BIM (building information modeling), a 3D/4D/5D representation of a construction job and an underlying database to query for information takes up the majority of the higher-end market, and has become more popular over about a fifteen-year period. Many of the large MEPF (mechanical/electrical/plumbing/fire suppression) subcontractors, representing 70% of the prime contract or more of a commercial building project, have gotten up-to-speed with how to use 3D software and collaborate on designs. The rest of the vendors and subcontractors that would also benefit from relying upon such a system are still left behind. For BIM to be more successful, it needs to have the following characteristics:
- It needs to be centralized, so that all stakeholders on a construction project can access timely information.
- It needs to be able to be updated on the fly, and be able to be queried quickly to figure out all quantity information.
- It needs to be accessible for all trades, in that additional software will not necessarily need to be purchased to access underlying project information.
- It needs to represent the other 30% of scopes on a project accurately, not just the MEPF trades.
Right now, BIM is just not accessible for every other stakeholder on a project. While the technology is rapidly increasing, the barrier of entry for other subcontractors, for example is still higher than necessary. 2D PDFs are still heavily relied upon for other sub-trades to estimate; the data is there, but it is just not shared properly.
Further, BIM should not only be used for higher-value projects, but also for smaller jobs. 3D scanning technology that also builds an accurate model through machine learning (currently being developed by Bentley) will need to create a model for existing spaces as accurately as possible before this can occur successfully.
Scanning and Re-Scanning
As large projects are underway, the re-scanning them at different intervals will be critical to understand if the design is being followed properly. This can happen by drones, or by other autonomous robots at night. IBM’s Watson platform can also be utilized to interpret data through AI; is a room completed or not? Can an automated punch list be put into place? There is a lot of data that would need to be stored and processed to make this kind of a system work. All parties having access to scanned data would allow for a more efficient build overall; project managers would not need to be in the field every day, and installers would not have to keep taking pictures to track progress.
The responsibility of scanning a site should be by the GC or CM; scanning should be sub-contracted out for several reasons:
- Insurance and risk mitigation.
- To avoid this task being repeated by every sub trade looking to cover themselves.
- For specialization purposes: The ability for a scanning subcontractor to increase their technical skills by doing dozens of projects a year. They will also be more knowledgable in the technology as they will be able to specialize in it, and know how to store/access data more effectively.
There may only be a handful of companies large enough to have their own internal scanning departments, just as they would have internal site surveyors for large civil jobs. For every other company, scanning should be an external function built into every contract.
As the construction industry has lagged behind others in productivity, there are numerous opportunities which can automate busy work by analyzing data and creating a greater access to information. Such systems that collect project data in a centralized location already exist. Procore, for example, keeps a log of construction documents, and has the ability to highlight revisions in sheets issued in addendum, amongst other tools. This is one great example of utilizing existing technologies to provide an upper hand for professionals in being able to sort through thousands of documents, and note key changes as a project progresses.
An initial planning phase to identify all of the important variables which may skew the costs of the project is critical in ensuring the project can be properly vetted. For example, a prevailing wage rate project might force a contractor to have to pay their installers more than they are regularly paid, and must be captured so that the costs are reflected in an estimate. Some of these important factors include:
- Estimated installation dates / project duration.
- Tax rate –not knowing this can throw billing off tremendously.
- Project delivery method (Design Build, Design Bid Build, CM-at-Risk, etc.)
- Effective work hours.
- Drug or background testing requirements.
- Badging requirements.
- Billing retention rate.
Many of these factors can be found in the 011000 Summary section of a specification book, in other areas of the front end of the spec book, and even the invitation to bid. Some must be gathered from the prime contract itself. Subcontractors may have to reach out to general contractors to fill in this list, but this information is critical in starting a new cost estimate. Not understanding these general issues will affect the performance of the project, and could eat significantly into margin if not caught. In a standardized system, the answers to these questions will have a place to be inputted, and their value will affect a generated cost estimate.
API stands for “Application Programming Interface”, and refers to software querying databases for specific information. An example in construction would be that if a company has a relationship with an outside vendor who maintains an eCommerce website, a program could potentially “ping” it for a price for a specific item or items under its own customer account (to reflect its own discount). The API could also calculate and return the necessary freight and minimum order fees for the item or items. When utilizing such an API, the vendor could ask for project information, such as industry, name, and location, to tie the product’s cost back to a formal quotation number. The use of an API in the case of estimating would help three-fold: 1) it would decrease the clutter of PDF files of formal quotations 2) it would decrease the amount of human interaction necessary to get cost information 3) cost information could be generated as the estimate is being performed
If there could be industry-wide standards on how API access could be implemented, we would be able to start to see workloads decrease in complexity. There is no standard in how a specification book should be written for example, between the government and many other private firms. A 1,500 page specification book does not serve any party effectively. There is also no standard in how updates to projects occur, and there are a dozen or more platforms for distributing files. Every platform must follow a standard; they can add or modify views as they wish, but there should be an underlying database that is shared for all stakeholders to access. Construction is still very much archaic compared to other industries; it does not have to be this way, but there needs to be an industry push for an equal working space.
Supply Chain Management
Supply chain management needs to be taken more seriously in commercial construction. Lead times need to be continually updated, not just from the accepting of a subcontract agreement, but throughout the submittal process all the way up to when the material arrives on site. Lead times exist and are constantly updated on vendors’ internal quoting websites for products or lines; a GC needs to be aware of changes automatically from the awarded subcontractor. Every item number needs to have the ability to be tracked. This is not occurring on a wider basis, and contributes to delays and general misinformation. Solving the supply chain and API issues would have the ability to save the industry billions of dollars per year.
One day, a task such as cost estimating a project may have the ability to go completely automated. Since my career has been heavily focused on cost estimation, I can confidently say that this will happen in the next 10-15 years. In fact the probably of my career being automated is currently at 57% likelihood. An estimator may become an employee who attempts to make sense of all of the generated project data, and may not need to perform takeoffs. They could act more as an engineer, monitoring the important parts of a project, instead of actively trying to make sense of it. Other roles in construction may do the same. There are many redundancies in construction because workers simply do not have the information that they need to do their jobs effectively. Until implemented, some of the described technologies will need to push the industry into that direction, where integrated project delivery is the norm and not the exception. The construction industry needs to come together to solve these large data and standardization problems soon, or we will continue to lag behind.