CHAPTER 9

Planning an Interdependent Project Portfolio

Sarah would have preferred more notice, but the overall decision to move into a new building has already been made, the plans are in progress, and the deadline is set. She can’t do much about it.

Her goals in planning are as follows.

Make sure the physical move happens on schedule without any data loss. (This includes ensuring that backups are completed and protected.)

Make sure new equipment is installed, tested, and operational by the deadline.

Keep critical department functions operating during the move.

Avoid crunch time as much as humanly possible.

DEFINITIONS

Here are the key terms you need in this chapter.

The control point identification chart is a tool for identifying problems to allow for early solution.

Finish-to-finish dependency describes a situation when the finish of task B is dependent upon the finish of task A. (It can start earlier.)

Frontloading is a strategy for doing as much of the project as you can up front. This builds in additional margin if things go wrong as you approach the deadline.

Lag time is extra time built into a project at the end to allow time to respond to emergencies. Unlike a lag activity, this time is not associated with any task; it can be used freely to respond to any emergencies.

Overlap dependency means the start of task B can begin sometime after the start of task A, but before task A finishes.

Start-to-start dependency is when the start of task B depends upon the start of task A. In a normal finish-to-start dependency, the start of task B depends on the finish of task A.

Task slack belongs to a single task, allowing it to be delayed without affecting the next task.

Path slack is slack that is shared among a sequence of tasks.

Free slack is the amount of time a task can be delayed before affecting the next task in the sequence.

Total slack is the amount of time a task can be delayed before affecting the deadline of the project.

PLANNING FOR INTERDEPENDENT PROJECT PORTFOLIOS

The process for planning the interdependent project portfolio is the same as that for the independent project portfolio.

Step 1: Define the Portfolio Goals

Make sure that you have a clear understanding of the portfolio goal and that you have prepared the objective in triple-constraint terms. You then need to determine the triple constraints for each of the projects in your portfolio—it is not necessarily the case that they will be the same.

In Sarah’s case, the goal is to have a successful move without losing data or production time. The triple constraints are as follows.

  Time: The move-in week has been scheduled for the week of September 22. All activities needed for the move-in must be ready by then. (Certain activities may take place after the move-in week, such as unpacking.)

  Performance: Critical computer services are not interrupted, data is not lost, new equipment is installed and operational, and the staff is trained.

  Budget: There is a fixed budget for new equipment ($1 million), and the overall move budget is $150,000.

Put the triple constraints in order. You can determine them by looking at the consequences of failure. The weak constraint is pretty obviously budget, since the damage done by going over budget is less than the damage done by missing the move-in or losing data and critical services.

Performance is probably the driver, because missing the move-in date, although costly and embarrassing, would be less serious than failing to deliver critical services or losing data. That gives us:

driver: performance

middle: time

weak: budget.

Remember, when you determine the triple constraints, you often look at the consequences of failure. This does not mean that you intend to fail or want to fail. In fact, you want to achieve all three constraints. You need to understand the order of the constraints so you can make good decisions in the event of unforeseen difficulties.

Step 2: Prepare a Work Breakdown Structure

This portfolio consists of four interrelated projects: 1) the physical move, 2) new equipment procurement, 3) continuing operations and transition management, and, 4) new center implementation (training, documenting, dry run, debugging) (see Figure 9.1). Those projects become level 2 of the work breakdown structure (WBS). Each project consists of one or more tasks, which go in level 3 of the WBS. Some tasks may be large and complex enough to be considered projects of their own; you may break them down further into levels of the WBS.

Use the WBS to plan the management of your project. First, you need to assign a manager to each of the projects in the portfolio.

Tip! Let the managers help you plan. This not only lowers the amount of work you have to do but also helps them get buy-in and acceptance of the workload.

Verify triple constraints of each project. For example, the physical-move project is time driven, but continuing operations is performance driven. New equipment may be performance driven, but it could be budget driven, depending on the organizational issues that affect your decisions.

Remember that individual project managers on your team manage their projects according to their goals. Your job is to ensure that their work fits into your overall strategy.

Step 3: Conduct the Task Analysis Process

When you analyze tasks at the level of an interdependent project portfolio, you need to think of it as an entire process, rather than just filling out forms. An enormous amount of strategic and tactical thinking takes place, and this is one of your chief opportunities to get the members of your senior project team properly oriented, and prepare them to get results.

Determine available resources. One of the frequent situations in managing an interdependent project portfolio is that the portfolio comes on top of a normal load of projects and work. Using the techniques in Section 1, you and your team need to determine what time you can realistically commit to the portfolio’s projects. This often means making decisions about what functions can be delayed until after completion of the portfolio. Delay everything that can be delayed, so you will have maximum time to accomplish what cannot be delayed.

Assign clear responsibilities. It is easy for important matters to slip in a project situation like this. People can easily get wrapped up in day-to-day activities. As the cartoon says, “When you are up to your rear end in alligators, it is hard to remind yourself that your original objective was to drain the swamp.”

To overcome this, make sure that each activity is in the hands of a specific person or team and that accountability and authority have been clearly defined.

Determine good enough. While each individual project in the interdependent portfolio has to be accomplished, the quality standards vary. Because individual project managers tend to maximize their projects without regard for the portfolio as a whole, one of your key jobs is to establish minimum and maximum performance standards for each project. It is best to do this as a team activity—to get buy-in from the project managers.

Why maximum standards? Each project takes resources from the total pool. If a project is satisfactory at a certain level and exceeding that level takes more resources, is the improvement worth the cost? You must determine this by contribution to total portfolio quality.

Brainstorm effective strategies. Use your team resources to do some extensive brainstorming when faced with an interdependent project portfolio. Look for ways to save time, improve quality, and lower cost. Look for ways to reinvent methods of doing the individual projects.

Determine ways to monitor progress. You will need a regular schedule of meetings and reports to stay on top of the process. Plan them in advance, and be sure to follow good meeting practice, such as always having an agenda.

One good tactic to both monitor and motivate progress is to make a wall chart. Take one large, empty wall and display PERT and Gantt charts on it. Mark all progress issues on the charts so that they are constantly in view of all staff.

Motivate early action. On any portfolio of this size and complexity, one key tactic is to get started early. While crunch time may be inevitable, you can at least minimize it if people start early.

Step 4: Prepare a PERT Chart for the Portfolio

Place the tasks from the WBS into a PERT-chart sequence. You inserted two milestones: Start and Project Finish. As with all milestones, they will have a time estimate of zero. They help organize your chart. One task, Move-in week, is a must-start activity; it is already scheduled and you need to work around it. Most tasks have to be done in time for the move-in; the move-in tasks are done during the move-in (they have a start-to-start dependency relationship).

Remember, the decisions you make in laying out your PERT chart are strategic decisions that affect your project. One of Sarah’s most important goals is to minimize the crunch time on this project. She followed two strategies in laying out her PERT chart.

1. Frontloading: The more of this project you can get out of the way before the move-in date approaches, the better off you are. When managing a deadline-sensitive project with the possibility of unforeseen complications, put as much as possible up front.

2. Build in lag: Schedule the project to allow as much free time as possible between the end of the tasks immediately preceding the move-in week and the move-in week itself. Internal lag time lowers risk and increases safety.

Step 5: Determine the Critical Path

Place time estimates on your PERT chart, and determine the longest (critical) path. Identify available slack time. Slack time has several benefits.

First, slack is a way to lower risk. If you use the optional PERT time-estimating formula and calculate sigma (σ) for the paths, try to have 2σ of slack and lag available. This gives you a base 95 percent chance of reaching the deadline on time. If 2σ is unrealistic, take all you can get.

Second, slack sometimes gives you the opportunity to optimize resources. If you have paths with excessive slack (more than 2σ), look to see if you can reassign people and resources from tasks on that path to critical tasks. The noncritical tasks take longer, but that is okay as long as you still have slack. The critical tasks go faster, which improves your total project performance.

Our computer-center PERT chart contains some special situations that might affect your planning (see Figure 9.3).

Task slack versus path slack. The task, Plan office move, contains one week of task slack. That means it can finish one week late without delaying the start of the next task or the end of the project. (This is also called free slack.) The entire task sequence from Plan equipment move through Develop operations plan (see Figure 9.3) is scheduled to take eleven weeks. The corresponding critical-path section (Needs analysis through Procure system) takes fourteen weeks. The difference, three weeks, is slack time.

Unlike the task slack situation, this three-week period belongs equally to all tasks. If Plan equipment move is late one week, it delays the start of Distribute timetable, which delays the start of Coordinate department priorities, which delays the start of Develop operations plan, but it does not delay the start of move-in week! In other words, all the tasks in that sequence can be a total of three weeks late (in any combination) before affecting the start of the next task.

The three weeks in this case are also called total slack. Plan equipment move has three weeks of total slack, which is how late it can be without affecting the critical path. It has no free slack because any delay in that task delays the start of the next task in the path. Develop operations plan has the same three weeks of total slack but also has three weeks of free slack—assuming all its predecessors finish on time!

Internal lag. Move-in week is a must-start activity (fixed in time, rather than just following its predecessor task), but that does not automatically mean that the project start date is fixed. Working backward from the move-in week task, the critical path is fourteen weeks to that point, which means that the project portfolio must start no later than the week of 6/16 to make the 9/22 deadline. However, this assumes that everything will go perfectly. A better strategy is to move the start date back. If you add two weeks of lag time to the critical path, backing the start date up to 6/2, you add that margin of safety to the portfolio (also lowering the crunch-time problem).

Figure 9.4 shows lag time as an activity. This is always a legitimate option, but it is not strictly necessary. Do it if it helps you keep better track of your project.

Step 6: Prepare a Gantt Chart for Calendar Management

In managing an interdependent portfolio, you should put greater emphasis on the PERT chart, because it does a much better job showing connections among the projects that make up the portfolio. As you mentioned, displaying a wall-sized PERT chart can be a powerful tool to help keep people focused on the goal.

A Gantt chart, however, does provide advantages in any time-sensitive project, because it shows the activities in a calendar form. There is no reason you must limit yourself to only one project view, especially if you are using project management software. Creating a Gantt chart from a PERT chart (or vice versa) is simply a matter of selecting the option from the menu and hitting Print.

In this chart, you chose not to show the lag as a separate activity; that is why there are no critical tasks in the project until you reach the move itself. Even the tasks you previously designated as critical can be late up to two weeks before they jeopardize the start of the move.

As in the independent project portfolio, you can use the Gantt chart as a way to monitor progress in calendar terms. You should also update the PERT chart to see progress in terms of dependency relationships.

Step 7: Prepare a Control Point Identification Chart for Critical, High-Risk Activities

The control point identification chart is a tool to help you identify something that might go wrong—or right—that would be out of the ordinary. Use a brainstorming session for each project in your portfolio to identify potential problems and opportunities. Then, identify early warning points, and list possible solutions and strategies.

While the planning issues for the interdependent project portfolio are similar to those of other projects, it is important that you adjust your perspective to the special issues that make this type of project different. By taking into account the differences in size and internal relationships, you will be well on the way to success.

EXERCISE

Analyzing Your Interdependent Portfolio

To manage your own interdependent project portfolio successfully, it is important that you start with the right focus. Make sure you know the answers to the following questions.

1. Describe the interdependent project portfolio that you have to manage. What is the ultimate outcome, and how will you measure when you have achieved it?
   • _______________________________________________________________________
   • _______________________________________________________________________
   • _______________________________________________________________________
   • _______________________________________________________________________
   • _______________________________________________________________________
2. What are the triple constraints of the portfolio, and how are they ranked in terms of priority?
   • _______________________________________________________________________
   • _______________________________________________________________________
   • _______________________________________________________________________
   • _______________________________________________________________________
   • _______________________________________________________________________
3. What resources do you have available (people, equipment, money)? How can you obtain additional resources if required? How can you use the flexibility in the weak constraint to help you meet your goals?
   • _______________________________________________________________________
   • _______________________________________________________________________
   • _______________________________________________________________________
   • _______________________________________________________________________
4. What skills, knowledge, and experience are needed in your team to accomplish your objective? What skills do you have? What skills can your team members supply? What skills, knowledge, and experience are missing? How will you get them?
   • _______________________________________________________________________
   • _______________________________________________________________________
   • _______________________________________________________________________
   • _______________________________________________________________________
   • _______________________________________________________________________
5. What political/organizational issues could affect your portfolio performance? What potential conflicts can you anticipate from team members and subordinate project managers? How do different people see your goal?
   • _______________________________________________________________________
   • _______________________________________________________________________
   • _______________________________________________________________________
   • _______________________________________________________________________
   • _______________________________________________________________________