Project schedule management is what most people think of when they think of project managers. It’s where the deadlines are set and met. It starts with figuring out what work you need to do, how you will do it, what resources you’ll use, and how long it will take. From there, it’s all about developing and controlling that schedule.
Rob and Rebecca have decided to tie the knot, but they don’t have much time to plan their wedding. They want the big day to be unforgettable. They want to invite a lot of people and show them all a great time.
But just thinking about all of the details involved is overwhelming. Somewhere around picking the paper for the invitations, the couple realize they need help…
Rob: We want everything to be perfect.
Rebecca: There is so much to do! Invitations, food, guests, music…
Rob: Oh no, we haven’t even booked the place.
Rebecca: And it’s all got to be done right. We can’t print the invitations until we have the menu planned. We can’t do the seating arrangements until we have the RSVPs. We aren’t sure what kind of band to get for the reception, or should it be a DJ? We’re just overwhelmed.
Rob: My sister said you really saved her wedding. I know she gave you over a year to plan.
Rebecca: But I’ve always dreamed of a June wedding, and I’m not willing to give that up. I know it’s late, but can you help us?
Since there are so many different people involved in making the wedding go smoothly, it takes a lot of planning to make sure that all of the work happens in the right order, gets done by the right people, and doesn’t take too long. That’s what the Project Schedule Management knowledge area is all about.
Initially, Kathleen was worried that she didn’t have enough time to make sure everything was done properly. But she knew that she had some powerful time management tools on her side when she took the job, and they’ll help her make sure that everything will work out fine.
You need to know the order of the Project Schedule Management processes for the exam. Luckily, they are pretty intuitive. Can you figure out the order?
The Plan Schedule Management process is just like all of the other planning processes you’ve seen so far. In fact, you’ve already seen all of the inputs and tools that are used to create it in previous processes. Just like with the Plan Scope Management process from Chapter 5, your goal is to build a Schedule Management plan from the other project management plans, your company’s culture and existing documents, and the project charter.
Define Activities uses everything we already know about the project to divide the work into activities that can be estimated. The inputs for this process all come from the processes in the Integration Management knowledge areas. The first step in Project Schedule Management is figuring out how the project work breaks down into activities—and that’s what the Define Activities process is for.
Kathleen wrote down everything she knew about the project. She used the activity list from her last wedding as a guide and then thought about the things that Rob and Rebecca wanted that were different from her past projects. She broke those things down into activities and pulled everything together into an activity list.
Sometimes you start a project without knowing a lot about the work that you’ll be doing later. Rolling wave planning lets you plan and schedule only the stuff that you know enough about to plan well.
If Kathleen were using rolling wave planning, she might write a schedule for only the tasks it takes to do the invitations, and leave the planning for the menu and the seating up in the air until she knows who will RSVP.
Rob and Rebecca probably wouldn’t be happy hearing that Kathleen was only going to plan for the invitations to be sent, though. They want to know that their wedding is going to happen on time. That’s why rolling wave planning should be used only in cases where it’s not possible to plan any other way.
Think back to the definition of a project in Chapter 2. Remember how projects are progressively elaborated? Rolling wave planning takes advantage of the fact that you know more about the project as you go to make plans more accurate.
Q: How would you use experts to help you define tasks?
A: A wedding is something that a lot of people have experience with, but some projects are not as easy to get a handle on. If you were asked to manage a project in a new domain, you might want to ask an expert in that field to help you understand what activities were going to be involved.
Even in Kathleen’s case, access to a catering expert might help her think of some activities that she wouldn’t have planned for on her own.
It could be that you create an activity list and then have the expert review it and suggest changes. Or, you could involve the expert from the very beginning and ask to have a Define Activities conversation with him before even making your first draft of the activity list.
Q: I still don’t get rolling wave planning.
A: One way to develop a project is to divide it up into phases of work, and gather requirements for each phase as the previous one is completed. Sometimes projects are done iteratively, where you divide the work up into phases and then plan out each phase before you execute on it. Rolling wave planning is all about committing to planning out one portion of the work that you’ll do, executing it, and then moving on to the next portion.
Software projects using agile methodologies use a form of rolling wave planning to make sure that everything they sign on to do gets done. They might do user stories for a release of the software up front, build it, and deliver it, and then gather more requirements based on the users’ ideas after working with the released version.
Here is part of a WBS. Arrange the activities underneath the WBS to show how the work items decompose into activities.
The main output of this process is the activity list. It’s the basis for all of the estimation and scheduling tasks you will do next. But there are a few other outputs that go along with it, and help to make the estimates more detailed and accurate.
This is a list of everything that needs to be done to complete your project. This list is lower-level than the WBS. It’s all the activities that must be accomplished to deliver the work packages. |
Here’s where the description of each activity is kept. All of the information you need to figure out the order of the work should be here, too. So any predecessor activities, successor activities, or constraints should be listed in the attributes, along with descriptions and any other information about resources or time that you need for planning. |
All of the important checkpoints of your project are tracked as milestones. Some of them could be listed in your contract as requirements of successful completion; some could just be significant points in the project that you want to keep track of. The milestone list needs to let everybody know which are required and which are not. NoteSome milestones for the wedding:
|
As the team is working through the work they’ll do, they sometimes find new information about the scope or requirements of the project. When that happens, they’ll need to create a change request to update the scope of the project. |
When the team identifies new activities there will likely be changes to both the schedule baseline and the cost baseline. You’ll need to put those changes through change control and then update the Project Management plan to include the new activities. |
Rob: The quartet cancelled. They had another wedding that day.
Rebecca: Aunt Laura is supposed to do the reading at the service, but after what happened at Uncle Stu’s funeral, I think I want someone else to do it.
Rob: Should we really have a pan flute player? I’m beginning to think it might be overkill.
Rebecca: Maybe we should hold off printing the invitations until this stuff is worked out.
Kathleen: OK, let’s think about exactly how we want to do this. We need to be sure about how we want the service to go before we do any more printing.
Now that we know what we have to do to make the wedding a success, we need to focus on the order of the work. Kathleen sat down with all of the activities she had defined for the wedding and decided to figure out exactly how they needed to happen. That’s where she used the Sequence Activities process.
The activity attributes and the activity list she had created had most of the predecessors and successors written in them. Her milestone list had major pieces of work written down, and there were a couple of changes to the scope she had discovered along the way that were approved and ready to go.
One way to visualize the way activities relate is to create a network diagram. Kathleen created this one to show how the activities involved in producing the invitations depend on one another.
For example, the calligrapher is the person who’s hired to write the addresses on the invitations, so Rob and Rebecca need to pick one before the invitations can be addressed. But the invitations also need to be printed before they can be addressed, because otherwise the calligrapher won’t have anything to write on! See how predecessors can get all complicated? Luckily, a diagram makes sense of them!
Showing the activities in rectangles and their relationships as arrows is called a precedence diagramming method (PDM).
Just looking at the way all of these tasks relate to one another can help you figure out what’s important at any time in the project. Once Rob and Rebecca looked at the network diagram below, they realized they needed to get online and start looking for a venue for their wedding right away, even before they’d figured out their budget and guest list.
The most common kind of dependency is the Finish to Start. It means that one task needs to be completed before another one can start. There are a few other kinds of dependencies, though. They can all be used in network diagrams to show the order of activities. The three main kinds of dependency are Finish to Start (FS), Start to Start (SS), and Finish to Finish (FF). This tool is called dependency determination and integration in the Sequence Activities process.
Sometimes your project will depend on things outside the work you are doing. For the wedding, we are depending on the wedding party before us to be out of the reception hall in time for us to decorate. The decoration of the reception hall then depends on that as an external dependency.
Rob and Rebecca really want the bridesmaids to arrive at the reception before the couple. There’s no necessity there—it’s just a matter of preference. For the exam, know that you should set discretionary dependencies based on your knowledge of the best practices for getting the job done.
You can’t address an invitation that hasn’t been printed yet. So, printing invitations is a mandatory dependency for addressing them. Mandatory dependencies are the kind that have to exist just because of the nature of the work.
The rehearsal dinner can’t begin until the happy couple leaves the church. Some dependencies are completely within the team’s control.
Sometimes you need to give some extra time between activities. Lag time is when you purposefully put a delay between the predecessor task and the successor. For example, when the bride and her father dance, the guests wait awhile before they join them.
Lead time is when you give a successor task some time to get started before the predecessor finishes. So you might want the caterer preparing dessert an hour before everybody is eating dinner.
Q: Where do you get the dependency information to figure out your network diagram?
A: Your activity attributes should list the predecessors and successors for each activity. As you build the network diagram, you might discover new dependencies as well. Your project team will determine the dependencies necessary for each of the activities.
Q: What about Start to Finish dependencies?
A: It’s possible for activities to require that a task has been started before it can finish. An example might be that singing couldn’t start until after the music had started. But tasks like that are pretty rare and almost never show up in network diagrams.
Q: My scheduling software makes network diagrams for me. Why do I need to know this?
A: Most scheduling software does create one of these diagrams automatically. But spending the time to think through your dependencies and examine them visually can really help you find places where you might need to give some tasks more priority if you want to get your project done on time. So you should know how to make them too.
Many people use applications like Microsoft Project to visualize their activity list and organize their dependencies. Taking the time to really understand the best sequence of activities can have a real impact on your project’s deadlines, so using software like that as part of your company’s project management information system is a good way to get a handle on just the right sequence for your project to run efficiently.
You should still think of things in sequence.
For the test, it’s important to know the order of these processes. And even though you might do it all at once, you probably spend some time thinking about each of these things.
As you sequence the activities, you will find new activities that need to be added and new attributes for activities that you didn’t know about. So, while the main product of this process is the network diagram, you also produce updates to some of the Define Activities documents and outputs of other processes, too.
Once you’re done with Sequence Activities, you’ve got everything you need to figure out how long each activity will take. That’s done in a process called Estimate Activity Durations. This is where you look at each activity in the activity list, consider the scope and the resources, and estimate how long it will take to perform.
Estimating the duration of an activity means starting with the information you have about that activity and the resources that are assigned to it, and then working with the project team to come up with an estimate. Most of the time you’ll start with a rough estimate and then refine it (maybe a few times!) to make it more accurate. You’ll use these tools and techniques to create the most accurate estimates.
Expert judgment will come from your project team members who are familiar with the work that has to be done. If you don’t get their opinion, then there’s a huge risk that your estimates will be wrong!
Parametric estimating means plugging data about your project into a formula, spreadsheet, database, or computer program that comes up with an estimate. The software or formula that you use for parametric estimating is built on a database of actual durations from past projects.
Bottom-up estimating means building your estimates up based on individual estimates from the people who will do the work.
Data analysis means using reserve analysis to add extra time to the schedule (called a contingency reserve or a buffer) to account for extra risk. It can also mean using alternatives analysis to think through all of the possible options to find the most efficient path for delivery.
Analogous estimating is when you look at activities from previous projects that were similar to this one and see how long it took to do similar work for them. But this only works if the activities and the project team are similar!
Three-point estimates are when you come up with three numbers: a most likely estimate that probably will happen, an optimistic one that represents the best-case scenario, and a pessimistic one that represents the worst-case scenario. The final estimate is the average.
Decision-making techniques help the team decide on the best estimates for the activities they’ve defined.
Meetings help teams to work together when they estimate the work.
You’ve got a list of activities, you know what resources are needed to actually do each activity, and you’ve got your estimation tools and techniques…now you have enough to create the estimates! That’s the whole point of the Estimate Activity Durations process, and it’s also the main output.
The activity duration estimates are estimates of how long each activity in the activity list will take. The estimate can be in hours, days, weeks…any work period is fine, and you’ll use different work periods for different jobs. A small job (like booking a DJ) may just take a few hours; a bigger job (like catering—including deciding on a menu, ordering ingredients, cooking food, and serving guests on the big day) could take days. You’ll also a produce a basis of estimates document that describes your approach, the assumptions you made, and factors you considered.
You don’t always know exactly how long an activity will take, so you might end up using a range (like 3 weeks +/- 2 days).
You’ll also learn more about the specific activities while you’re estimating them. That’s something that always happens—you have to really think through all of the aspects of a task in order to estimate it. So the other output of Estimate Activity Durations is updates to the project documents.
You may have guessed from the name that the activity duration estimates are always duration estimates, not effort estimates, so they show you calendar time and not just person-hours.
The activity duration estimate consists of estimates for each activity. It’s the main output of the Estimate Activity Durations process.
Q: When you use parametric estimation, how does the program or formula know how much to estimate?
A: When people design a system for parametric estimation, they collect a lot of data from past projects and condense it into a table or a database. And then they come up with a heuristic (like a rule of thumb) that lets you boil your estimation down into just a few parameters that you need to enter. Most successful parametric estimation systems need a lot of time to develop.
Q: Since reserve analysis lets me use buffers, why can’t I just put everything I don’t know about into the reserve?
A: The idea behind reserve analysis is that there are always unknowns on any project, but you can account for these unknowns by taking your best guess at what’s going to go wrong and inserting a buffer. But you can’t just make an enormous reserve, because then there’s no reason to ever do any estimation! The entire project becomes one big unknown, and that’s not particularly useful to anyone.
Q: Wait a minute! I don’t quite get the difference between a duration estimate and an effort estimate. Can you explain?
A: Duration is the amount of time that an activity takes, while effort is the total number of person-hours expended. If it takes two people six hours to carve the ice sculpture for the centerpiece of a wedding, the duration is six hours. But since two people worked on it for the whole time, it took 12 person-hours of effort to create!
Kathleen’s really got a handle on how long things are going to take, but that’s not enough to get the job done. She’s still got some work to do before she’s got the whole project under control.
The Develop Schedule process is the core of Project Schedule Management. It’s the process where you put it all together—where you take everything you’ve done so far and combine it into one final schedule for the whole project. A lot of project managers consider this the most important part of their job. The schedule is your most important tool for managing a project.
Develop Schedule brings everything you’ve done so far together into one final schedule. All of the outputs from the other Project Schedule Management processes are inputs to Develop Schedule.
Don’t worry, even though you’re done with the Estimate Activity Durations process, you’re not done with the resources.
You’re never going to have the complete resource picture until you’re done building the schedule. And the same goes for your activity list and duration estimates, too! It’s only when you lay out the schedule that you’ll figure out that some of your activities and durations didn’t quite work. We’ll go in depth about resources when we talk about Project Resource Management in Chapter 9.
That’s why the processes have the word “Estimating” in their names! Because you’re taking an educated guess, but you won’t know for sure until you’ve actually developed the schedule.
Each of the processes allows updates to an output from a previous one, so when you discover changes, you can include them in the schedule.
Rebecca: Well, let’s see. What menu did we give to the caterers?
Rob: We didn’t give it to them yet, because we won’t have the final menu until everyone RSVPs and lets us know which entrée they want.
Rebecca: But they can’t RSVP because we haven’t sent out the invitations! What’s holding that up?
Rob: We’re still waiting to get them back from the printer. We can’t send them out if we don’t have them yet!
Rebecca: Oh no! I still have to tell the printer what to print on the invitations, and what paper to use.
Rob: But you were waiting on that until we finished the guest list.
Rebecca: What a mess!
…but it turns out to be a lot bigger than either Rob or Rebecca realized at first! How’d a question about one guest’s meal lead to such a huge mess?
The critical path method is an important tool for keeping your projects on track. Every network diagram has something called the critical path. It’s the string of activities that, if you add up all of the durations, is longer than any other path through the network. It usually starts with the first activity in the network and usually ends with the last one.
The reason that the critical path is, well, critical, is that every single activity on the path must finish on time in order for the project to come in on time. A delay in any one of the critical path activities will cause the entire project to be delayed.
The critical path is the string of activities that will delay the whole project if any one of them is delayed.
Knowing where your critical path is can give you a lot of freedom. If you know an activity is not on the critical path, then you know a delay in that activity may not necessarily delay the project.
This can really help you handle emergency situations. Even better, it means that if you need to bring your project in earlier, you know that adding resources to the critical path will be much more effective than adding them elsewhere.
It’s easy to find the critical path in any project! With a little practice, you’ll get the hang of it. Of course, on a large project with dozens or hundreds of tasks, you’ll probably use software like Microsoft Project to find the critical path for you. But when it does, it’s following the same exact steps that you’ll follow here.
Start with an activity network diagram.
Find all of the paths in the diagram. A path is any string of activities that goes from the start of the project to the end.
Find the duration of each path by adding up the durations of each of the activities on the path.
The critical path is the one with the longest duration!
The goal is to find the float for each activity. We’re not really concerned with finding a total float for each path—we’re looking at the activities independently.
Once you’ve figured out the critical path, there’s all sorts of useful stuff you can do with it. One of the most useful things you can do is calculate the float. The float for any activity is the amount of time that it can slip before it causes your project to be delayed. You might also see the word slack—it’s the same thing.
Luckily, it’s not hard to figure out the float for any activity in a network diagram. First you write down the list of all of the paths in the diagram, and you identify the critical path. The float for every activity in the critical path is zero.
There are three paths in this network:
Start → A → B → C → Finish = 11
The path with the longest duration is the critical path!
Start → D → E → Finish = 7
Start → D → F → G → Finish = 8
The float for each of the activities on the critical path is zero.
Find the next longest path. Subtract its duration from the duration of the critical path, and that’s the float for each activity on it.
That’s how you figure out how long any of its activities can slip before it delays the project.
Do the same for the next longest path, and so on through the rest of the network diagram. Pretty soon, you’ll fill in the float for every activity!
This is why you figure out the length of each path before you start.
You can use this method to find the float for every activity in a network diagram. Another word for float is slack.
Once you know the float, you know how much play you have in your schedule. If an activity has a float of 2 days, it can slip by that much without affecting the end date.
To find the float for an activity, figure out how much it can slip before it makes the project late. The float for any activity on the critical path is ZERO!
Q: Where do the duration numbers come from on each activity?
A: A lot of people ask that question. It’s easy to forget that everything you do in Sequence Activities builds on the stuff you did in the other Project Schedule Management processes. Remember the estimates that you came up with during Estimate Activity Durations? You used techniques like three-point estimates, analogous estimating, and parametric estimating to come up with an estimate for each activity. Those are the estimates that you use on your network diagrams!
Q: What if there’s a path that’s not critical, but where even a small slip in one activity would delay the project?
A: This is exactly why it’s important to know the float for each of your activities. When you’re managing your project, it’s not enough to just pay attention to the activities on the critical path. You need to look for any activity with a low float. And don’t forget that there may be some activities that aren’t on the critical path but still have a float of zero! These are the ones where you really want to pay attention and watch out for potential resource problems.
All of the processes in Project Schedule Management tie together! When you develop your schedule, you’re using the durations for your activities that you came up with in Estimate Activity Durations.
Coming up with the float for each activity is useful, but you can actually do better! When you have a long critical path, but the other paths in your network diagram are short, then you have a lot of freedom in when you can start and finish each of the activities that are not on the critical path. You can use early start and early finish to get a handle on exactly how much freedom you have in your schedule.
Is the earliest time that an activity can start. An activity near the end of the path will only start early if all of the previous activities in the path also started early. If one of the previous activities in the path slips, that will push it out.
Is the earliest time that an activity can finish. It’s the date that an activity will finish if all of the previous activities started early and none of them slipped.
When you find the early start and early finish for each task, you know exactly how much freedom you have to move the start dates for those activities around without causing problems.
It’s also important to know how late any activity can run before it delays the project. That’s what late start and late finish are for! They let you figure out how late you can start a certain task and how much it can slip before it delays your project.
Is the latest time that an activity can start. If an activity is on a path that’s much shorter than the critical path, then it can start very late without delaying the project—but those delays will add up quickly if other activities on its path also slip!
Is the latest time that an activity can finish. If an activity is on a short path and all of the other activities on that path start and finish early, then it can finish very late without causing the project to be late.
Figuring out the late start and late finish will help you see how much “play” you have in your schedule. An activity with a large late start or late finish means you have more options.
You can use a method called forward pass to add the early start and finish to each path in your network diagram. Once you’ve done that, you can use backward pass to add the late start and finish. It makes your network diagrams look a little more complicated, but it gives you a lot of valuable information.
Take a forward pass through the network diagram. Start at the beginning of the critical path and move forward through each activity. Follow these three steps to figure out the early start and early finish!
The ES (early start) of the first activity in the path is 1. The EF (early finish) of any task is its ES plus its duration minus 1. So start with Activity A. It’s the first in the path, so ES = 1, and EF = 1 + 6 – 1 = 6.
Now move forward to the next activity in the path, which is Activity B in this diagram. To figure out ES, take the EF of the previous task and add 1. So for Activity B, you can calculate ES = 6 + 1 = 7, and EF = 7 + 5 – 1 = 11.
Uh-oh! Activity C has two predecessors. Which one do you use to calculate EF? Since C can’t start until both B and D are done, use the one with the latest EF. That means you need to figure out the EF of Activity D (its ES is 1, so its EF is 1 + 2 – 1 = 2). Now you can move forward to Activity C and calculate its EF. The EF of Activity D is 2, which is smaller than B’s EF of 11, so for Activity C the ES = 11 + 1 = 12, and EF = 12 + 7 – 1 = 18.
You can use a backward pass through the same network diagram to figure out the late finish and start for each activity.
The backward pass is just as easy as the forward pass. Start at the end of the path you just took a pass through and work your way backward to figure out the late start and finish.
Start with the critical path. You’re calculating the latest any activity can start and finish, so it makes sense that you need to start at the end of the project and work backward—and the last activity on the critical path is always the last one in the project. Then do these three steps, working backward to the next longest path, then the next longest, and so on, until you’ve filled in the LS and LF for all of the activities. Fill in the LF and LS for the activities on each path, but don’t replace any LF or LS you’ve already calculated.
Start at the end of the path, with Activity C. The LF (late finish) of the last activity is the same as the EF. Calculate its LS (late start) by subtracting its duration from the LF and adding 1. LS = 18 – 7 + 1 = 12.
Now move backward to the previous activity in the path—in this case, Activity B. Its LF is the LS of Activity C minus 1, so LF = 12 – 1 = 11. Calculate its LS in the same way as step 1: LS = 11 – 5 + 1 = 7.
Now do the same for Activity A. LF is the LS for Activity B minus 1, so LF = 7 – 1 = 6. And LS is LF minus duration plus 1, so LF = 6 – 6 + 1 = 1.
Now you can move onto the next longest path, Start-D-C-Finish. If there were more paths, you’d then move on to the next longest one, and so on, filling in LF and LS for any nodes that haven’t already been filled in.
When a task is a predecessor to two later tasks, use the one with the lower LS to calculate its LF.
All of this critical path stuff seems pretty serious, right? It seems like one of the toughest concepts on the exam...at first! But don’t sweat it, because it’s actually not hard—it just takes a little practice. Once you do it yourself, you’ll see that there’s really nothing to worry about.
Calculating the ES, EF, LS, and LF may seem complicated, but it only takes a little practice to get the hang of it. Once you walk through it step by step, you’ll see that it’s actually pretty easy!
You won’t have to do this kind of thing on the job…that’s what computers are for!
Project management software like Microsoft Project will do these calculations for you. But you need to know how to do it yourself, because when the computer is doing critical path analysis, this is exactly how it figures it out!
Q: Would I really use this critical path stuff in real life, or is it just something I need to memorize for the PMP exam?
A: Yes, critical path analysis really is important in real life! Sure, for a small project with a dozen or so activities, it’s pretty easy to figure out which activities are critical and which can slip by a little bit. But what happens if you’ve got a project with dozens of team members and hundreds of activities? That’s where critical path analysis can come in very handy. For a project like that, you’d probably be using project management software rather than calculating the critical path yourself, and the software will be able to highlight that path for you. Pay special attention to all of the activities that are on the critical path—those are the ones that could potentially delay the project.
Q: What about the other numbers? How do I use float?
A: Float is a very powerful planning tool that you can use to figure out how well your project is going, and to predict where your trouble spots might be. Any activity with a low or zero float absolutely must come in on time, while the people performing an activity with a larger float have more freedom to slip without delaying the project. So you might want to assign your “superstar” resources to the low-float activities, and those people who need a little more mentoring to the ones with higher float.
Q: OK, but what about late start, early finish, and those other numbers? Do those do me any good?
A: Early and late start and finish numbers are also very useful. How many times have you been in a situation where you’ve been asked, “If we absolutely had to have this in two months, can we do it?” Or, “How late can this project realistically be?” Now you can use these numbers to give you real answers, with actual evidence to back them up.
Here’s an example. Let’s say you’ve got an activity in the middle of your project, and one of your team members wants to plan a vacation right at the time that the activity will start. Do you need to find someone to fill in for him? If he’ll be back before the late start date, then your project won’t be late! But that comes at a cost—you’ll have used up the extra slack in the schedule.
Q: I can see how the critical path is useful on its own, but what does it have to do with the rest of Project Schedule Management?
A: If you start putting together your schedule but the activities are in the wrong order, that’s really going to cause serious problems…and sometimes doing critical path analysis is the only way you’ll really figure out that you’ve made that particular mistake. That’s why you need to pay a lot of attention to the Sequence Activities tools and techniques. If you’ve come up with an inefficient or inaccurate sequence, with too many or incorrect predecessors and dependencies, then your entire critical path analysis will be useless.
There are two important schedule compression techniques that you can use to bring in your project’s milestone dates…but each has its own cost. When you absolutely have to meet the date and you are running behind, you can sometimes find ways to do activities more quickly by adding more resources to critical path tasks. That’s called crashing.
Crashing the schedule means adding resources or moving them around to shorten it. Crashing ALWAYS costs more and doesn’t always work!
Then you can’t crash the schedule.
There’s no way to crash a schedule without raising the overall cost of the project. So, if the budget is fixed and you don’t have any extra money to spend, you can’t use this technique.
Another schedule compression technique is called fast-tracking. Sometimes you’ve got two activities planned to occur in sequence, but you can actually do them at the same time. On a software project, you might do both your user acceptance testing and your functional testing at the same time, for example. This is pretty risky, though. There’s a good chance you will need to redo some of the work you have done concurrently.
On the exam, if you see something about “overlapping activities,” it’s talking about fast-tracking.
Crashing and fast-tracking are SCHEDULE COMPRESSION tools.
It’s always a good idea to think about all of the things that could go wrong on your project in advance. Trying to think through all of the possible problems your project could run into is called what-if analysis.
What if the limo breaks down?
What if the florist cancels at the last minute?
What if the dress doesn’t fit?
What if the band gets sick?
What if the guests get food poisoning?
What if there’s a typo in the church address on the invitation?
What if the bridesmaids don’t show up?
What if the cake tastes horrible?
What if we lose the rings?
That way, you can figure out how to deal with any problems that might come your way. Sometimes there’s no way to still meet your dates and deal with these scenarios. But it always makes sense to try to understand the impact they will have on your schedule.
This is a specific kind of what-if analysis where you model uncertainty using a computer. There are some packages that will help to calculate risk using random numbers and Monte Carlo analysis algorithms. While this is not a commonly used technique, there might be a question or two about it on the PMP exam, and you should know what it is.
Using a project management software package to create a model of the schedule and adjusting various elements to see what might happen is another technique for analyzing network diagrams.
There are just a few more tools and techniques in the Develop Schedule process that you should know.
Sometimes only one resource can do a given activity. If that resource is busy doing another activity on the critical path, the path itself needs to change to include that dependency. That’s the point of resource leveling. It evaluates all of the resources to see if the critical path needs to change to accommodate resource assignments.
If you made any mistakes in your leads and your lags, you might be able to adjust them to change the projected end date.
The last two tools and techniques in the Develop Schedule process are the ones you just learned over the last few pages: schedule compression and schedule network analysis using critical path, float, and the other schedule analysis techniques you just learned.
Agile teams typically have a cadence or timebox they use to develop. An Agile release plan takes the list of user stories that have been estimated for a release and uses their estimated size of effort to map them to an upcoming iteration. By presenting the release plan in this way, the team gives the product owner the ability to reprioritize stories and change the order in which stories will be released.
Your company might have scheduling software that they use to generate schedules. You might also use a project management information system to reference existing company calendar information that might impact your schedule.
Of course, the main product of Develop Schedule is the schedule. But there are a few other supporting documents that help you understand how the work will get done as well.
All of that analysis and modeling should produce a schedule that everyone can get behind. After thinking your way through everything that can go wrong and assigning resources, you should have a pretty accurate prediction of the work required to complete the project.
These are the same change requests that you’ve seen in other processes. They should be really familiar by now!
Calendars will help you keep track of the time when team members will be away on vacation or unavailable to work on your project.
The schedule data is a collection of information about your schedule. It will include things that you’ll need to analyze your schedule later on in the project: alternative schedules, specific requirements for resources, milestone charts, bar charts, project schedule network diagrams, and other data and metrics about your schedule.
Before you can do change control, you need requested changes. Once the change is approved, you can update the baseline.
When the Develop Schedule process is complete, a baseline is created so that actual progress can be compared to the plan.
Since the schedule baseline and Schedule Management plan are both part of the Project Management plan, it makes sense that it would have to be updated.
While you’re creating the schedule, you might find that you need to update calendars, your resource requirements, the attributes of the activities themselves, or even your risk register, to name a few possibilities.
Q: Don’t we need to go through change control before we update the resource requirements or the activity attributes?
A: No. You need to go through change control if you are requesting changes to, say, your Cost Management plan. But while you are working on creating your schedule, everything you have created as part of the Project Schedule Management knowledge area is fair game.
As you work your way through your network diagram and figure out new dependencies, you will find that you need more resources for some items or that the activity itself has changed. That’s why this process gives you the freedom to refine your earlier idea and make all of the Project Schedule Management documents sync with your new understanding.
The Develop Schedule process is about taking all of the information you can think of up front and putting it into a schedule that is realistic. When you are done with this process, you should have a really good idea of what you are going to do, who will do it, and how long it will take.
Q: We always want to do our projects as quickly as we can. Why don’t we always fast-track and crash our schedules?
A: Because crashing is expensive and fast-tracking is risky. While it may look good on paper to add a lot of resources to a project that is running late, it often adds so much management overhead and training issues that the project just comes in later.
Even though it might seem like some predecessors are really unnecessary, you usually planned them for a reason. So when you break your dependencies to fast-track your project, you can significantly compromise the quality of the work that gets done. That means you might have to redo it altogether—which would probably take a lot of time.
While fast-tracking and crashing might work sometimes, they always add both risk and cost to your project.
Q: Do people really do Monte Carlo analysis to figure out their schedules? I have never heard of that before.
A: It’s true that most people don’t use this technique to figure out what might go wrong on their projects, so don’t feel bad if you’ve never heard of it. Some people think that this is just one of those things that is on the PMP exam, so you have to know what it is. But there really are some project managers who use it and get great results!
Updates refine the outputs of previous processes so you don’t have to go back and redo them.
Kathleen doesn’t just sit around and wait for schedule changes to happen…
Joe (on phone): Good afternoon, Joe’s Catering. Joe speaking. How can I help you?
Kathleen: Hello, Joe. This is Kathleen calling about Rob and Rebecca’s wedding.
Joe: Oh, hi! Everything’s going fine with that wedding.
Kathleen: Are you sure? What about that big convention across town that’s going to be happening at the same time? Won’t it be tough to find waiters in June?
By realizing that the convention across town will need waiters, too, Kathleen prevents a lot of changes before they cause schedule problems!
Joe: I didn’t think of that; we’d better start figuring out how we’ll handle it now.
The project manager doesn’t just wait for change to happen! She finds the things that cause change and influences them.
As the project work is happening, you can always discover new information that makes you reevaluate your plan, and use the Control Schedule process to make the changes. The inputs to Control Schedule cover the various ways you can discover that information. The outputs are the changes themselves.
All of the documents you made during the Develop Schedule process get updated during the Control Schedule process. Here’s a closer look at what those updates mean.
As your project is happening, you might need to revise your list of assumptions to include new information about activity sequencing or duration.
You may add supporting details to estimates—like assumptions that were made—to help people understand how the team came up with the estimate. As you make changes to your project schedule, you should be tracking your lessons learned so that other projects can benefit from your experience. Sometimes you might find changes to your basis of estimates as your project progresses.
Throughout your project, you’ll update the project schedule with any new information you learn as you execute your plan.
Changes to resource calendars and risks need to be updated as you go. You might find that resources who originally planned to be on vacation end up changing their plans, or you might identify new risks as you go.
Schedule data is information about the schedule, like start date, end date, and information about the critical path. In some cases, things change so drastically while your project is executing that you need to create new activities and forecasted duration dates to get an accurate picture of your schedule.
Managing schedule change means keeping all of your project documents up to date.
Most often, you identify changes by looking at performance data. It’s just as important once you make a change to gather performance data as it was when you found the change in the first place. Here’s how performance data feeds into the Control Schedule process.
The Control Schedule process turns work performance data into work performance information.
The tools and techniques for Control Schedule are all about figuring out where your project schedule stands. By comparing your actual project to the schedule you laid out in the baseline and looking at how people are performing, you can figure out how to handle every schedule change.
Data analysis
In this process you use data analysis to evaluate the information coming in from the project to forecast whether or not your team is going to meet its forecasted dates. There are two important calculations called schedule variance (SV) and schedule performance index (SPI) that give you valuable information about how your project is doing. Those two calculations are part of the earned value technique, which you’ll learn all about in the next chapter.
Agile teams often use iteration burndown charts to gauge whether they are ahead or behind on work they’ve committed to accomplishing within a given timeboxed iteration.
Performance reviews are used to compare actual dates versus the dates in the schedule baseline. Trend analysis lets you see if performance of the team is getting better or worse over time. Variance analysis is when you dig in and try to understand the cause of the variance between actual dates and the dates in the baseline. What-if scenario analysis allows you to model possible schedule changes to mitigate risks or respond to changes.
Leads and lags, and schedule compression
Most of the tools from the last process apply to this one too. As you find variances in the schedule, you need to figure out the impact of those issues and change your schedule to account for the new information.
Resource optimization
As things change in your project, you need to make sure that resources are covering all of the activities in your plan. That means you need to distribute resources so that the work that needs to get done always has a resource available to do it.
Critical path method
Use the precedence diagramming method to figure out your early start and late finish for all of the tasks in your project. Then calculate the float to find which tasks are in the critical path. Once you know the critical path, you can choose to fast-track or crash the schedule to optimize the amount of time your project will take.
Project management information system
This is software like Microsoft Project that helps you organize and analyze all of the information you need to evaluate the schedule of any project.
Q: When I create work performance information, who uses it?
A: The work performance information that you create is used by a lot of people. The team uses it to keep an eye on the project. If there’s a schedule problem coming up, it alerts the team so that they can help you figure out how to avoid it.
Performance information is also used by your project’s sponsor and stakeholders, who are very interested in whether or not your project is on track. That information gives them a good picture of how the project is doing…and that’s especially important in Control Schedule, because most change control systems require that every change is approved by a change control board that includes sponsors and stakeholders.
Q: What’s schedule data used for?
A: You use the schedule data to build the schedule, and you’ll usually generate and analyze it using a schedule tool (like Microsoft Project). It includes detailed information about things like resource requirements, alternate best-case and worst-case schedules, and contingency reserves.
When you put together your schedule, you should look at all of these things in order to create an accurate plan. The more information you have when you’re building your schedule, the more likely it is that you’ll catch those little problems that add up to big schedule slips.
Q: One of the tools is project management software. Do I need to know how to use software in order to pass the exam?
A: No. The PMP exam does not require that you know how to use software like Microsoft Project. However, if you spend a lot of time using project management software, then you probably have become very familiar with a lot of the Project Schedule Management concepts. It’s a good way to learn the basics of schedule management.
Q: How often am I supposed to update the project calendar?
A: The project calendar shows you the working days for your team, holidays, nonworking days, planned training, and the dates that could affect your project. Luckily, in most companies these dates don’t change very often. You probably won’t need to update it—and most project managers just use their company’s existing project calendar.
When you’re doing Develop Schedule, you may discover that you need to make a change to the project calendar. That’s why updates to the project calendar are an output of Develop Schedule.
Q: What do I do with work performance data and work performance information once I’ve collected it?
A: When you’re planning your project, you’ll often look to your company’s past projects to see what went well and what could have been planned better. And where do you look? That information is in the organizational process assets.
So where do you think that information comes from? It comes from project managers like you who added their work performance data and information to the company’s organizational process asset library.
Any time you generate data about your project, you should add it to your organizational process assets so you can use it for future projects.
You’ll see change control over and over again—every single knowledge area has its own change control process! Luckily, you’ll start to see how similar they all are. But Control Schedule has its own quirks, and they’re important for understanding Project Schedule Management.
Answers in “Control Schedule Magnets Answers”.
Rob and Rebecca had a beautiful wedding! Everything was perfect. The guests were served their meals, the band was just right, and everyone had a blast…
You’ll see change control over and over again—every single knowledge area has its own change control process! Luckily, you’ll start to see how similar they all are. But Control Schedule has its own quirks, and they’re important for understanding Project Schedule Management.
You’re managing a project when your client tells you that an external problem happened, and now you have to meet an earlier deadline. Your supervisor heard that in a situation like this, you can use schedule compression by either crashing or fast-tracking the schedule, but he’s not sure which is which. What do you tell him?
Crashing the project adds risk, while fast-tracking adds cost.
When you crash a project, it always shortens the total duration of the project.
Crashing the project adds cost, while fast-tracking adds risk.
When you fast-track a project, it always shortens the total duration of the project.
Given this portion of the network diagram to the right, what’s the EF of activity F?
41
49
53
61
Given this portion of the network diagram to the right, what’s the LS of activity F?
41
49
53
61
You are managing a software project. Your QA manager tells you that you need to plan to have her team start their test planning activity so that it finishes just before testing begins. But other than that, she says it can start as late in the project as necessary. What’s the relationship between the test planning activity and the testing activity?
Start-to-Start (SS)
Start-to-Finish (SF)
Finish-to-Start (FS)
Finish-to-Finish (FF)
You’re managing an industrial design project. You’ve come up with the complete activity list, created network diagrams, assigned resources to each activity, and estimated their durations. What’s the next thing that you do?
Use rolling wave planning to compensate for the fact that you don’t have complete information.
Create the schedule.
Consult the project scope statement and perform Sequence Activities.
Use fast-tracking to reduce the total duration.
Which of the following is NOT an input to Develop Schedule?
Activity list
Project schedule network diagrams
Resource calendars
Schedule baseline
Three members of your project team want to pad their estimates because they believe there are certain risks that might materialize. What is the BEST way to handle this situation?
Estimate the activities honestly, and then use a contingency reserve to cover any unexpected costs.
Allow more time for the work by adding a buffer to every activity in the schedule.
Tell the team members not to worry about it, and if the schedule is wrong it’s OK for the project to be late.
Crash the schedule.
You’re managing a software project. You’ve created the schedule, and you need to figure out which activities cannot slip. You’ve done critical path analysis, identifying the critical path and calculating the early start and early finish for each activity. Which activities cannot slip without making the project late?
The ones with the biggest difference between ES and LF
The activities on the critical path
The activity with the most lag
The last activity in the project, because it has no float
What is the critical path in the activity list to the right?
Start-A-B-C-Finish
Start-A-D-E-F-Finish
Start-G-H-I-J-Finish
Start-A-B-J-Finish
Name | Predecessor | Duration |
---|---|---|
Start | — | — |
A | Start | 6 |
B | A | 4 |
C | B | 8 |
D | A | 1 |
E | D | 1 |
F | E | 2 |
G | Start | 3 |
H | G | 3 |
I | H | 2 |
J | B, I | 3 |
Finish | F, J, C | — |
What is the float for activity F in the activity list to the right?
0
7
8
10
You’re managing an interior decoration project when you find out that you need to get it done earlier than originally planned. You decide to fast-track the project. This means:
Starting the project sooner and working overtime
Assigning more people to the tasks at a greater total cost, especially for activities on the critical path
Starting activities earlier and overlapping them more, which will cost more and could add risks
Shortening the durations of the activities and asking people to work overtime to accommodate that
Slack is a synonym for:
Float
Lag
Buffer
Reserve
You’re managing a construction project. You’ve decomposed work packages into activities, and your client needs a duration estimate for each activity that you come up with. Which of the following will you use for this?
Milestone list
Activity list
Critical path analysis
Project schedule network diagram
What’s the correct order of the Project Schedule Management planning processes?
Sequence Activities, Define Activities, Estimate Activity Durations, Develop Schedule
Plan Schedule Management, Define Activities, Sequence Activities, Develop Schedule, Estimate Activity Durations
Plan Schedule Management, Define Activities, Sequence Activities, Estimate Activity Durations, Develop Schedule
Plan Schedule Management, Develop Schedule, Define Activities, Sequence Activities, Estimate Activity Durations
Which of the following is NOT a tool or technique used in Estimate Activity Durations?
SWAG estimation
Parametric estimation
Analogous estimation
Three-point estimation
You’re managing a project to build a new project management information system. You work with the team to come up with an estimate of 27 weeks. In the best case, this could be shortened by two weeks because you can reuse a previous component. But there’s a risk that a vendor delay could cause the project to be delayed by five weeks. Use PERT to calculate a three-point estimate for this project.
25.83 weeks
26 weeks
27.5 weeks
28.3 weeks
Given the network diagram below, what’s the critical path?
Start-A-B-C-End
Start-A-D-G-End
Start-E-D-C-End
Start-E-F-G-End
For that same network diagram below, what’s the float for activity A?
0 weeks
1 week
2 weeks
4 weeks
For that same network diagram below, what’s the float for activity E?
0 weeks
1 week
2 weeks
4 weeks
You’re managing a software project when your customer informs you that a schedule change is necessary. Which is the BEST thing to do?
Consult the schedule management plan.
Notify the team and the sponsor that there’s going to be a schedule change.
Influence the factors that cause change.
Refuse to make the change because there’s already a schedule baseline.
Your company has previously run other projects similar to the one you’re currently managing. What is the BEST way to use that information?
Check the organizational process assets for lessons learned and other information about the past projects.
Use parametric estimation to estimate your project based on past projects’ performance.
Start from scratch because you don’t want mistakes from past projects to influence you.
Reuse the Project Management plan from a past project.
You’re planning the schedule for a highway construction project, but the final date you came up with will run into the next budget year. The state comes up with capital from a reserve fund, and now you can increase the budget for your resources. What’s the BEST way to compress the schedule?
Go back to your three-point estimates and use the most optimistic ones.
Use the extra budget to increase your contingency reserve.
Hire more experts to use expert judgment so your estimates are more accurate.
Crash the schedule.
You’re managing a construction project. You’ve decomposed work packages into activities, and your client needs a duration estimate for each activity that you came up with. Which of the following BEST describes what you are doing?
Evaluating each activity to figure out how much effort it will take
Estimating the number of person-hours that will be required for each activity
Understanding, in calendar time, how long each activity will take
Estimating how many people it will take to perform each activity
Answer: C
You’re likely to get some questions that ask you about crashing and fast-tracking, and it’s important to know the difference between them. When you crash the project, it means that you add resources to it, especially to the critical path. There’s no real risk in doing that—in the worst-case scenario, the extra people just sit around!—but it does cost more. Fast-tracking means adjusting the schedule so that activities overlap. The same resources are doing the work, so it’s not going to cost more, but it’s definitely riskier, because now you’ve eliminated buffers and possibly broken some dependencies! And remember that crashing or fast-tracking won’t always work to make the project go faster!
Answer: C
To calculate EF for a task with two predecessors, start by calculating the ES: choose the highest EF of the predecessors and add 1: ES = 40 + 1 = 41. Then use that ES to calculate EF by adding the duration and subtracting 1: EF = 41 + 13 – 1 = 53.
Answer: B
To calculate LS for a task that’s the predecssor to two other tasks, start by calculating the LF. Take the lowest LS and subtract 1: LF = 61 – 1 = 61. Then use that LF to calculate LS by subtracting the duration and adding 1: LS = 61 – 13 + 1 = 49.
Answer: C
Don’t let the jargon fool you! You don’t need to know anything about software testing to answer this question. When you have two activities, and the first activity has to be timed so that it finishes before the second one starts, then you’ve got a Finish-to-Start relationship, or FS.
Answer: B
Did answer A trick you? No need for rolling wave planning when you’ve got enough info to define all the activities!
This is a which-is-next question that describes a project that’s completed the Define Activities, Sequence Activities, Estimate Activity Resources, and Estimate Activity Durations processes. The next process in Project Schedule Management is Develop Schedule, which means that the next thing you do is create the schedule!
Answer: D
The schedule baseline is an output of the Develop Schedule process, not an input. You should definitely know what goes into the schedule baseline: it’s a specific version of the schedule that you set aside and use for comparison later on, when you want to know if the project is running late.
Answer: A
You always want to be honest with your estimates. Every project has unknowns and risks, and there’s no way to estimate any activity exactly! Luckily, we have tools to deal with this. You can use reserve analysis, a tool of Estimate Activity Durations, to come up with a contingency reserve that you can use to plan for these risks.
Answer: B
The critical path is the path in the network diagram where any delay will cause a delay in the schedule. These are the activities that cannot slip without making the project late.
Answer: A
When you draw out a network diagram for the activities in the table, you end up with four paths. And you definitely should draw out the activity diagram for a question like this! You’re allowed to use scratch paper on the exam, and this is one place where you should definitely do it. Of the four paths, only one has the longest duration: Start-A-B-C-Finish, which has a duration of 6 + 4 + 8 = 18. That’s the critical path.
Activity F is in the path Start-A-D-E-F-Finish. This path has a duration of 6 + 1 + 1 + 2 = 10. The float of an activity is the longest time it can slip before it affects the critical path. In this case, activity F can slip by 8 without causing the path that it’s on to go beyond the critical path. But any more than that, and its path becomes the new critical path!
Did you notice answer A? Don’t forget that the float of any activity in the critical path is zero!
Answer: C
This is the definition of fast-tracking, and you’re probably getting the hang of this one by now. You may get a question like this, but you’ll almost certainly see fast-tracking as an incorrect answer to several questions.
Answer: A
Remember that when you see slack, it’s the same thing as float. Either term could appear on the exam.
Answer: B
This question is asking about the Estimate Activity Durations process. Take a look at the answers—there’s only one answer that’s used in that process: you need to start with the activity list in order to do the estimates for the activities. The other answers are things that are inputs, tools, or techniques for other processes.
When a question asks what you’d use for a process, it’s asking you to pick an input, tool, or technique that’s part of the process.
Answer: C
It’s not hard to remember the order in which the Project Schedule Management processes are performed. If you use a little common sense, you can reason your way through a question like this. You need to figure out your planning procedures before you can define your activities, you need to define your activities before you can sequence them, and you need to do all of that before you can build a schedule!
Control Schedule isn’t included in the list of processes because if a schedule change happens, you’ll have to go back and revisit the other Project Schedule Management processes. So it doesn’t have a specific order!
Answer: A
You’ll have to know the different kinds of estimating techniques for the exam. You don’t necessarily have to be good at doing them, but you should recognize which are which. Parametric estimating is when you plug values into a formula, program, or spreadsheet and get an estimate. Analogous estimating uses similar activities from past projects to calculate new estimates. Three-point estimating uses an optimistic, pessimistic, and realistic estimate.
Answer: C
This question is asking you to apply the PERT three-point estimation formula: (optimistic time + 4 × most likely time + pessimistic time) ÷ 6. When a question gives you these values directly, it’s easy. But in this case, to answer the question you had to figure out the values for the optimistic time and pessimistic time, which meant that you needed to look at the assumptions that the team was making. The most likely time was given: 27 weeks. The best-case scenario would come in two weeks earlier, at 25 weeks, and the worst case would come in five weeks late, at 32 weeks. So the estimate is (25 weeks + 4 × (27 weeks) + 32 weeks) ÷ 6 = 27.5 weeks.
Sometimes you’ll get a question about applying a formula, but you’ll need to read the text in the question to figure out all of the variables.
Answer: C
The path Start-E-D-C-End has a duration of 8 + 2 + 4 = 14, which is the longest total duration in the entire network.
Answer: B
Activity A is on three different paths: Start-A-B-C-End (13), Start-A-D-C-End (12), and Start-A-D-G-End (9). To calculate its float, you take the longest path’s length and subtract it from the length of the critical path: 14 – 13 = 1.
Answer: A
Can you think of how a question might quiz you on this information without actually asking you to look at a network diagram?
Since Activity E is on the critical path, its float is zero, because the float of any activity on the critical path is zero.
Answer: A
The Schedule Management plan tells you how changes to the schedule are to be handled. Any time there’s a change, the first thing you should do is consult the plan to see how it should be handled.
Answer: A
The organizational process assets contain historical information about past projects. When you write up your lessons learned, or create work performance information, you store it in your company’s organizational process asset library! Also, did you notice that answer B was the wrong definition of parametric estimation?
Answer: D
Crashing the schedule is the form of schedule compression that increases cost. This is a difficult question because all of the answers sound good, and one or two are a little misleading! Don’t fall into the trap of choosing an answer because you recognize a valid tool or technique in it. Reserve analysis and three-point estimates are very useful techniques, but they’re not the answer to this question.
Answer: C
This question was really about the definition of duration, and the key to answering it is to understand how duration is different from effort. The correct answer talks about “calendar time,” which is what a duration is: it’s a measurement (or estimate) of how long the activity will take in real life, taking into account the number of people who will be doing the work, the availability of the people and other resources, everyone’s vacation time, time taken away from the schedule because people are pulled off of the activity to work on higher-priority activities, and other real-world factors. That’s different from effort (which is often measured in person-hours), and it’s different from resource estimating (which involves estimating how many people and what other resources will be used for the activity).