WHY SCHEDULE STANDARDIZATION IS CRITICAL FOR AI

The advantage of using Work Packages and LEAN Look-Ahead Schedules

We’re entering the next revolutionary age, the age of Artificial Intelligence. And it has the potential to finally lead our industry to see higher productivity and profits that other industries have already seen with the use of technology. But our industry still has a significant hurdle to overcome - digitization and standardization. The good news is, there’s more than likely an app these days that can replace any of your paper-based solutions. This can easily solve the digitization issue, but it still does not solve the standardization issue.

When it comes to extracting intelligence from data by applying AI, the data is the fuel. And one might think that you could simply throw all your data into the magical AI black box and it’ll analyze it and give you intelligent insights. But unfortunately, and fortunately (robots won't replace us!) it still needs humans to clean the data, especially with complex systems like in construction projects. Companies that identify their company goals can quickly zero-in on the data sets, and structure it to allow AI to give better insights. Another good thing when it comes to standardization - our industry has plenty of processes that can be followed, such as MasterFormat, Uniformity, WBS, Construction Work Packages (CWP), etc.

If you’re a large construction management firm, you’re probably already using some or all of these, and that's great. But from my meetings with several Top ENR contractors and smaller GCs, I’ve discovered that most companies need a little fine-tuning. You can have all the strings on a guitar but each one needs to be tuned to produce the desired sound, and each chord needs to also be in the right place. This is when a company needs to identify their goals and fine-tune their process and procedures but needs to be extremely careful to not make redundant steps that create no value. In a study, The Boston Consulting Group found that “over the past fifteen years, the number of procedures, vertical layers, interface structures, coordination bodies, and decision approvals needed…has increased by anywhere from 50 percent to 350 percent” and “managers spend 40 percent of their time writing reports and 30 percent to 60 percent of it in coordination meetings.”

So how can you standardize schedules to make them more accurate and realistic for future projects? It’s actually pretty simple, by using the existing processes of the Master Schedule and Construction Work Packages, and integrating them with the Look-Ahead Schedule.

Master Schedule

Simply put, it’s a list of high-level activities logically outlining how the sequence of work will be carried out from the start of the project to the end of the project with intermediate goals highlighted in between, known as milestones. The level of detail should account for major trades onsite to know who goes first and who directly follows to ensure proper hand-offs of work from start-to-finish. For example, a line item that might be seen in the master schedule could be erecting the structural steel of the superstructure or MEP blockouts, but you can probably get away with not putting in install shower curtain rod because no major tasks are depending on it other than hanging a shower curtain. But each team will need to decide this on their own. I typically tell clients if it’s something important, has a large value associated with it, or has other critical tasks tied to it, it probably needs to be in the schedule. It goes back to the old phrase, “You can’t manage what you don’t measure.” Just remember the plan is not just for your team, it needs to bring enough transparency for all shareholders, such as subcontractors or the design team, to have a general understanding of the project, so that they have a better understanding on how they need to execute their work and which dates are critical for the project.

The next common question that I get asked is, “Should I put the procurement tasks into the schedule?” This goes back to what I said earlier - if it’s a task that has a lot of tasks tied to it in the future, then yes. For example, submittals, scopes of work, and permits should be put in the schedule because in this day and age work cannot take place onsite unless a subcontractor has a signed contract and the work they plan on doing has approved shop drawings/submittal.

One of the most critical things when it comes to creating a master schedule is how to name a task. It seems simple, but I have seen a lot of schedules with just one word and no parent task to describe the location or task itself. For example, with a one-word description like “paint”, “install”, “concrete”, etc. it’s impossible to apply valuable AI to this or for even human intelligence (a person) to read or accurately figure out the task. Properly naming a task is important because it gives AI a way to analyze the task and start seeing actual patterns and compare it with other schedules or best practices in work sequences that you have internally built.

Example Broken Down by System

> Summary Task: <System>-<Location>

Example: A.1.0-Curtain Wall-Floor 20

> Subtask A.1.1-<Action Word>-<Scope Call to Action>-<Specific Location>-< Trade>

Example: Install Curtain Wall Type 1 East Elevation GL5-10 - Enclose

Example Broken Down by Location

> Summary Task:<Location>-<System>

Example: A.2.0-<Floor 20-Interior Metal Framing>

> Subtask A.1.0-<Action Word>-< Scope Call to Action >-<Specific Location>-< Trade>

Example: Layout Interior Units Walls Zone A – Monday Drywall

Construction Work Packages (CWP)

CWPs are more common in the ECP, oil and gas industries. But it’s slowly being adopted in commercial and residential construction. It’s defined as “the smallest unit of a Work Breakdown Structure. When preparing a Work Breakdown Structure using the decomposition technique, deliverables are generally broken down into smaller, more manageable chunks of work.” The graphic below shows how the CWP fits into traditional project/work breakdown structures, and after reading and looking at the graph, I bet most project managers would say to use common sense or that it’s not necessary to go into much detail. But this is when I think back to what my old platoon sergeant preached, “piss poor planning equals piss poor performance.”

The other good thing about CWP is that it allows handoffs between procurement, construction and field crews to go more smoothly. Most of the information for a CWP can be answered by the procurement team but should be reviewed and confirmed by the construction team once the CWP is ready to be performed out on the jobsite.

Just keep in mind that the CWP needs to deconstruct a task from the master schedule to explicitly give a deliverable package to a specific trade of workers and that packages should assist the construction management team in estimating the cost, duration, and required resources (workforce, materials, equipment) to complete the task.

Helpful Tips:

> Don't over-deconstruct. To figure out an appropriate work package size, try the 8/80 rule, which states that a work package shouldn’t take less than eight hours to complete or more than 80 hours.

> Make sure it’s measurable. Consider whether the work package can be completed between reporting periods.

> Keep estimates in mind. Remember that your breakdown should help you to estimate time and money for the project.

> Assign responsibility. For reporting and management purposes, a work package should be assigned to one person.

> Make it unique. The work package should not be repeated elsewhere in the WBS - it should be one of a kind.

Look-Ahead Schedules

An LAS breaks down and fine tunes the work that will take place over the next day and/or weeks with the goal of increasing the probability of each task finishing on time. The work that you planned on performing this week is typically known as the Weekly Work Plan and should have all the constraints removed and be ready to be executed. If there are still constraints such as RFIs, change orders, shop drawings, etc., then move it to the backlog and focus on the next activities with the highest priority. An LAS in a perfect world could be a copied and pasted week-to-week from the master schedule, and you could just add a little more detail. However, each task needs to be analyzed based on the CWP and LAS guidelines to increase the probability of its completion.

> Guidelines and functions of the Look Ahead Process:

> Form the workflow sequence and production rate to meet the schedule durations.

> Match labor and resources with planned work.

> Establish specific start and finish dates assigned to each crew to be able to communicate proper handoffs from one trade to another.

> Deconstruct the master schedule activities and finalize CWP.

> Go through constructability maze of how each trade’s work will be performed and then determine how it will be a sequence as a whole with the other trade.

> Where there are high interdependencies, figure out the best way to increase the resiliency.

> Assess if special tools or equipment are needed.

> Check for any missing information, such as RFIs or submittals.

> Is there a lay-down area or large space required for materials?

> Know what backlog of ready-work could be executed if something planned can no longer be performed.

> Update the master schedule from the look-ahead.

> Record actual start dates for activities started.

> Record accurate percentage completed.

> Provide root-cause or lesson learned on activities that didn’t go as planned.

The Takeaway: Fine-tune your schedules so that AI can optimize your future projects

The construction companies that understand the power of technology and standardization of data will be the ones to stay ahead of their competition. When it comes to outdated methods, the phrase “we’ve always done it this way” could soon lead you on the same path as Kodak, RCA or Blackberry. These are just a few companies that did not adapt or innovate fast enough and ignored the signs. But with simple implementations of standardized data, AI can be your secret weapon to higher productivity and profits.