Construction Scheduling. Schedule Performance Measurement, Part 3. (For Owners).

When monitoring and responding to the Contractor’s schedule performance on the project, how do you know if the schedule update is even valid for use?

When we review a schedule progress update, we need to first determine if the schedule update is a valid update. By that, I mean is the logic still complete? Have there been any revisions with this update that result in the schedule not meeting the contract requirements or best practices? It doesn’t take much to render a schedule useless….

There are most likely logic revisions, changes to lag values, or possible changes to calendars or assignments. There are also deletions and additions of activities and/or Activity Relationships. Then there are changes to quantities or Resources which may impact Activity Durations or Resource Calendars.

For this post, we will address changes to Relationship Lags.

Personally, the first schedule “quality check” I make is for complete logic. Please see the previous post, Schedule Performance Measurement, Part 2. (For Owners).

One of the things that are really hard to check, using Primavera P6, is changes to Activity Relationship Lags.

Often, when the Contractor is making schedule revisions to recover lost time for the update period, they make subtle changes to logic. They also make changes to the Relationship Lag values assigned to specific Activity Relationships. If they have modeled the sequence of work using SS Relationships with Lags, they may reduce the Lag value for an activity, driven by the Activity Relationship with the Lag, to gain a day or so. If there is an FS Relationship with a Lag, they may reduce that lag value as well.

They can also change the Relationship Lag’s effect by changing the Activity Calendar of the Activity selected to determine the Relationship Lag’s behavior.

None of this is easy to check in Primavera P6 and is not readily transparent in a Gantt Chart.

Most of us use third-party software designed to identify changes to Relationships and Relationship Lags. That’s great, but we still need to understand the impact of the changes identified.

Changes to Relationship Lag values does not make the schedule invalid for use. But, the owner needs to know what changes were made and what effect the changes have on the affected Logic Path. Some specifications prohibit the use of Relationship Lags. Many specifications allow limited use and do not allow the use of negative Lags at all. I prefer to minimize the use of Lags and do not like to use Negative Lags. There are conditions which warrant the use of a Negative Lag, but the condition has to be monitored and other options for modeling the work sequence should be exhausted.

That said, finding the changes to Relationship Lags, understanding the impact of the changes found, and reporting those findings is important to the Owner.

As  Planning and Schedule Professionals, we need to provide the Project Team with the “tools” they need to manage the project. The owner needs to know what the performance of the actual work in the field is, compared to the planned performance. Unless we have a valid schedule update to measure with, and against, and understand the impact of acceptable schedule revisions, we cannot provide an accurate measurement.

I’m sure many of you have comments or additional insight into this subject. Please share!

I’d love to hear what you think!

Please visit to learn more about Construction and Schedule Management Services, LLC

Please visit my LinkedIn account to learn more about me.

Please visit my “The Blue Book” ProView.

Paul Epperson CCM, PMP, PSP, PMI-SP

Construction Scheduling. Use of Relationship Lags.

When your contractor develops the project baseline schedule, do they use many relationship lags? Do you know?

A relationship lag is an imposed period of time between the predecessor and successor activity. A common use in a Start-to-Start (SS) relationship with a 2-day lag. Perhaps the contractor’s plan is to start finishing drywall 2 days after the drywall installation begins. This allows the work to run concurrently.

There absolutely are times when work can and should be modeled this way. However, care should be taken to ensure an open start isn’t inadvertently created. (This is a future topic)

The problem arises when the series of work activities is developed with SS relationships with an assigned lag. This may indicate the work is not developed to enough detail to allow Finish-to-Start (FS) relationships to drive the work. The issue is, once the start of the predecessor is actualized, the start of the successor is only dependent on the calendar set for use for lags. This may not match the number of days intended. If this work happens to fall on the Longest Path, this path is now driven by a calendar. I prefer to limit the use of SS relationships in general, much less with assigned an assigned lag.

Then there are the Finish-to-Finish (FF) relationships with lags. These allow the successor work to finish only after the predecessor’s finish date plus the lag duration. This is commonly used to close off SS relationships to model concurrent work with the finish dependency. However, often the only successor is the FF relationship with the lag. This creates an issue in itself. (This is a future topic)

The most serious issue is when the relationship is an FS with a lag. What does the lag model? Should there be an activity to establish this difference in the relationship finish to start? Most likely, the answer is yes. There is nothing wrong in using an FS with a lag to have the lag represent say, cure concrete. I personally prefer to add the cure concrete activity with the number of days required. This is more transparent. The lag is not visible and creates a non-work period seen in the Gantt Chart bars and can cause confusion.

Then there is the negative lag, or leads. Many schedulers like negative lags and use them as a viable way to model the plan. When I see an FS relationship with a negative lag, I wonder why the use of a SS relationship with a positive lag could not have been used. The use of the negative lag may better represent the thought that the successor cannot start until 3 days prior to the finish of the predecessor. And this 3 days is the critical period. I personally prefer to model this type of plan with a SS with a lag or to further break down the activities to allow the use of FS relationships to model the work.

Finally, there are the SS relationships with an assigned lag that is larger than the activity duration forcing the model to show the work as an FS relationship. Why not just use the FS relationship? Unless there is a reason to model the work this way, it just doesn’t make sense to do so.

The relationship lag, if used correctly, can allow the schedule to model the plan. It is my opinion that the use of lags should be kept to the absolute minimum and only used to model SS or FF relationships.

But remember, the lag follows the calendar set for use for lag calculation. If the predecessor calendar is set for use, and the predecessor is assigned to a 5-day work week calendar, the lag will follow that calendar and the duration will be in work days. So, a 3-day lag could become a 5-day delay between the relationships if the lag spans a weekend. This may not have been the case when project work was planned and the schedule was developed and the lag did not span a weekend. Remember, this is not transparent, only someone who understands the schedule software and has access to the native file format can verify this. Unless the schedule layout includes the columns which show relationships with lags and the column that shows calendars. Even then, you need to know what the actual non-work days for the specific calendar are.

I know many of you can offer additional comments and recommendations. I welcome your comments and input. My goal, as always, is to help our industry and help the projects we support….

I’d love to hear what you think!

Please visit to learn more about Construction and Schedule Management Services, LLC

Please visit my LinkedIn account to learn more about me.

Please visit my “The Blue Book” ProView.

Paul Epperson CCM, PMP, PSP, PMI-SP

Construction Scheduling. The problem with Start to Start Relationships. Or, Why doesn’t a Delay to this Activity Duration push my Finish Date?

person-question-300x300Do you regularly use Start-to-Start SS relationships? Perhaps you add a lag to drive the start of the successor out a few days?

Have you had problems with this practice? Perhaps a delay to the completion of the predecessor activity is actualized, but does not impact the schedule?

This is a common practice, unfortunately. The best practice is to add an additional successor with a Finish-to-Start FS relationship, or even a Finish-to-Finish FF relationship. This will provide the logic that allows the delay to the completion of the activity to drive other work. Without that additional successor activity, what you have is an “open end” and the activity finish does drive any successive work. It could never complete and you would not see the problem in the logic.

This condition is simple to prevent. Just make it a practice to always add a FS or FF successor when using a SS relationship. The problem usually happens when you’re trying to recover time and start to schedule work concurrently. You simply forget to add the additional successor….

Personally, I like to develop my schedules with all FS relationships unless the plan actually calls for a SS condition. If I have the need to use SS relationships, I assume my work isn’t broken down into enough detail to allow me to sequence the work logically.

There are always “exceptions to the rule”, but this is one I try to follow.

What other methods have you found to successfully utilize SS relationships, with or without lags?

I’d love to hear what you think!

Please visit to learn more about Construction and Schedule Management Services, LLC

Please visit my LinkedIn account to learn more about me.

Please visit my “The Blue Book” ProView.

Paul Epperson CCM, PMP, PSP, PMI-SP

Schedule Management. How should a Non-Schedule Savvy PM or Superintendent Develop and Manage a “Recovery Schedule”?

Struct SteelYou have a baseline schedule, with a few periodic updates completed, and now you need to make a revision to the schedule to dramatically re-sequence work, add/delete work to/from the project, or provide a required “Recovery Schedule”. What do you do?

This is one of the more difficult schedule management tasks. In some ways, it is almost more difficult than creating the baseline schedule.

Depending on the amount of time you need to “recover”, you will need to find the root causation for the schedule slippage and get that under control.  This could mean anything from having the party responsible for the slippage work additional hours to maintain the daily scheduled productivity, adding resources to increase the daily scheduled productivity, or resequencing of work to mitigate the lack of daily scheduled productivity or model concurrent work.

Next you will need to develop the most cost-effective means of accelerating work on the current longest path. This is tricky. Your schedule may have a near critical path which is very close to the scheduled longest path. As you shave days off the longest path, it will shift to the near critical path making that path the longest path. You will need to keep working at accelerating activities on the longest path, as it shifts until you reach the point of “recovery”.

That’s the easy part.

Now you have to obtain support for this plan from the project team. Subcontractors have to agree to provide what is needed to achieve this revised plan. Deliveries must be verified. Resource availability must be verified. There is probably additional cost involved. This must be managed a well.

In summary, any “recovery plan” will most likely involve concurrent work and/or acceleration of work. In any case, submitting a “recovery schedule” without the support of the project team for execution of the revised plan is a disservice to the owner, the project team and the project. You must be able to gain the commitment of the project team.

Schedule management includes many tasks and processes. Developing the “recovery schedule” is one of the more difficult tasks. The Project Manager should always be intimately involved with the planning of the revised work plan and “buy-in” from the project team is necessary for the successful execution of the “recovery plan”.

You may be able to manage the development of the recovery schedule without any problems.

However, when in doubt, seek out the advice of a professional planner and scheduler.

Please visit to learn more about basic schedule concepts.

Please visit my LinkedIn account to learn more about me.

Paul Epperson CCM, PMP, PSP, PMI-SP

Construction Scheduling Terms. What is the “Data Date”? What is “OOS”? What are “Lags”?

Pert P3You are working with a construction scheduling consultant and they use terms you’ve heard before, but you’re not 100% sure you really understand what they mean.  More importantly, you’re not sure how they affect the schedule.

Do not be ashamed, why should you have any in depth knowledge of yet another specialized function of the project management process? That’s why professional planning and scheduling consultants exist!

What is the “Data Date” and why do I care?

The data date is simply the “as-of” date used for progressing the periodic schedule update.

Typically, the data date is the end of the month, but I’ve seen it be the 15th and the 25th. Each schedule specification is different based on the owner’s requirements for reporting. Note: MS Project uses the term “Status Date”.

This is the date that all progress should be updated for progress up to. Everything after the data date is the pan to execute the remaining project activities. There should never be work in progress or incomplete with dates prior to the data date. I’ve seen this all too often with schedules run in MS Project because the person doing the schedule update does not understand this concept and how to adjust the settings to force incomplete work to move to the status date.  It’s a simple fix for an experienced scheduler. Always look at your schedules and know the data date and verify all work prior to the data date is complete and all work after the data date shows no progress. Think of the data date as the starting off point for the rest of your project…. This is a necessary practice for construction scheduling.

What is “OOS” and why do I care?

OOS or Out-of-Sequence is simply a way of saying that work actually completed did not progress as scheduled.

Typically, this involves an activity that was scheduled to start immediately after the finish of it’s “predecessor” having actually started prior to it’s “predecessor” finishing. Other reasons may be work progressing much differently than scheduled or work being suspended for whatever reason.

Why do you care? You care because this logic is still driving any “successor” activities and is not correct based on how you actually completed the work. It affects all downstream work logically tied to the finish of the OOS work and does not model the remaining work correctly.

This can affect change order fragnet insertions, not produce correct total float values, and does not provide you with an accurate as-built schedule. Best practice is to correct the logic for OOS work immediately after updating the progress each period update. Most schedule consultants do this, but many do not and most PM’s and Superintendents do not. Many owners are not aware of the need to correct OOS relationships and do not require or enforce this practice. This is unfortunate. There are times y0u can leave some OOS relationships in place, but absolutely OOS work on the longest path or near critical path should be correct. Most construction scheduling consultants correct OOS relationships each schedule update.

What are “Lags” and why do I care?

Lags are like invisible activities added to the schedule; which typically separate start dates of successor activities by a set amount.

Depending on which calendars are assigned and the relationship selected, a current activity’s successor activity will start immediately after the finish of the current activity, if the relationship is a Finish-to-Start, FS relationship. Adding a lag of 3 to this relationship would force the start of the successor to wait 3 work periods until starting.  Sometimes this is used for concrete curing time…… I prefer to use an activity for this so the time is transparent.

But typically, lags are used with Start-to –Start, SS relationships to allow work to be schedule concurrently. The lag forces the successor activity to wait set number of work periods after the start of the predecessor activity starts before starting. This models a plan to start one piece of work and then start successive work a set duration later. Such as finishing gypsum board wall sheathing before the entire area covered by the activity to hang the gypsum wallboard has completed. This is fine if this represents your plan to execute this work. Unfortunately, this is also how most PM’s and Superintendents choose to get the scheduled finish date back on time after the periodic update shows the project is behind schedule. This is still fine if the work can actually be executed this way, and it is the actual plan to “recover” lost time. It is just so darn easy to do, that many times it is used to appease the owner only and no plan is used to actually accomplish the work concurrently. This happens when the schedule is only part of the invoice process, and not an integral part of the project management process.

Don’t be afraid of lags, but the use of lags should be judicious and justifiable. There are many ways to model the logic to reduce the number of lags used. Typically, breaking the work into smaller pieces will allow the work to be scheduled without the use of lags.

Negative lags are bad idea and should be avoided or disallowed. There are many ways to make the logic work without the use of negative lags. Most schedule consultants stay away from negative lags.

MS Project shows lags and relationship types other than Finish-to-Start, FS in the predecessor and successor columns. Primavera P6 does not. Although it is not always easy to understand the effect a lag is having on the schedule, you should always be aware of their usage and understand the effect they have on the project schedule.

When in doubt, seek out the advice of a professional planner and scheduler.

Please visit to learn more about basic schedule concepts.

Please visit my LinkedIn account to learn more about me.

Paul Epperson CCM, PMP, PSP, PMI-SP