Topic 3: Project time management
The schedule is 'the conversion of a project action plan into an operational timetable' (Meredith & Mantel 2009: 333).
The PMBOK Guide process related to schedule or project time management comprises:
* define activities
* sequence activities
* estimate activity resources
* estimate activity durations
* develop schedule
* control schedule.
Project schedule management tools and techniques
Effectively, a schedule is nothing more than a date or time to perform an activity. Typical schedule presentation formats used by project managers include a milestone table or accomplishments , a 'Gantt chart' (which is called by many names including a 'bar chart' and PERT [Project Evaluation and Review Technique] chart (which is also called by many names, including a 'logic chart').
Each activity should be thought of as a process step that requires some specific, clearly defined items (inputs) to be available before it can start performs some sort of process to these items and delivers another clearly defined set of products (outputs).
An entire project's outputs are understood, the project may be decomposed in two ways: by product (outputs) and by development phase. This process of design–build–integrate is referred to as the system development life cycle or product development life cycle. The top-down decomposition, and bottom-up scheduling is known as 'top-down planning, bottom-up estimation'.
A process is typically a fine-tuned set of clearly defined steps that is repeated often. A project is simply a once-off process to achieve a specific outcome. The key difference between the two is that a project puts duration on each activity after the logical sequence has been determined.
If one activity produces an output at its end (finish) that is required to be in place before the second activity can start, then the dependency is a finish–start dependency. If other activities can start as soon as, or shortly after, their predecessor activity it is a start–start dependency. Start–start (or SS) dependencies tend to use lags more than the finish–start dependencies do. Leads, lags and non-finish–start relationships are important to understand when we create complex schedules. However, use them sparingly as techniques tend to reduce the time available between activities and, in so doing, also tend to eliminate any 'buffers' in the schedule.
Activities are scheduled (time-phased) based upon two simple and key principles:
1. that the materials must be available to undertake the work
2. the person(s) who will do the work will be available.
Resource levelling or resource smoothing occurs where the manager assigns a sequence to the activities merely to enable them to be done with the range of resources that are available; this is also called a discretionary dependency. Mandatory and discretionary dependencies control how a project manager assigns their resources to complete the project's activities.
When a number of activities are being undertaken, and some are sequential and some are parallel, we refer to the project as having a range of paths to completion, or multiple paths. The critical path is the sequence of activities (path) that determines (drives) the final end date of the project; it is the path that takes the longest time to complete is called the critical path; the critical path is the sequence of activities (path) that determines (drives) the final end date of the project.
Float or slack is the amount of time that a task in a project network can be delayed without causing a delay to:
* subsequent tasks ("free float")
* project completion date ("total float")
An activity on critical path has "zero free float", but an activity that has zero free float may not be on the critical path. Total float is associated with the path. If a project network chart/diagram has 4 non-critical paths then that project would have 4 total float values. The total float of a path is the combined free float values of all activities in a path.
The total float represents the schedule flexibility and can also be measured by subtracting early dates from late dates of path completion. Float is core to critical path method
This leads to two important concepts:
1. The only way to shorten the project duration is to shorten the duration of the critical path. Reducing the duration of other paths will not effect the end date.
2. Once the duration of the critical path is reduced to a certain amount, it ceases to be the critical path and another path is likely to be the critical path. Therefore, any further reduction in the duration of the old critical path will have no further effect on the project end date.
The extent to which an activity may be delayed before making an impact to the project end date is called the float. By definition, the critical path has zero float. Early start schedules have their float at the end, late start schedules have their float at the beginning, resource-levelled schedules uses the float to reduce resource requirements.
Project managers have found that they don't have to complete all of the schedule in order to begin the work. This 'just-in-time' scheduling is known as rolling wave scheduling. In a rolling wave plan, the finer detail is not known, but the overall approach, staffing, cost, duration and risk is well known and understood.
In most examples, the activity has been shown as a box, with arrows between the activities depicting the dependencies or precedence between the activities. This is sometimes referred to as activity-on-node (AON) where a node is a point on the diagram where the various lines (dependencies) come into or come out of. Another approach called activity-on-arrow (AOA), which start at one milestone, and point to the next milestone.
The critical chain method is intended to account for the behaviour of the project team. It recognises that project team members are more efficient if they have only one activity to focus on, and that the project team may be able to work harder than they are working now. The critical chain method determines mandatory logic as normal, and it undertakes resource assignment and levelling almost as normal. Activity durations are reduced by 20–40%, and the resultant savings put into buffers at the end of each path or chain, as against having float associated with each individual task (tasks that are not on the critical path).
Project managers have found that they don't have to complete all of the schedule in order to begin the work. Provided that the high-level plan is defined, and that sufficient detail is available to estimate the milestone dates and overall budget and resource utilisation—then more detailed planning can occur just prior to the work being undertaken. In a rolling wave plan, the finer detail is not known, but the overall approach, staffing, cost, duration and risk is well known and understood. Rolling wave scheduling is used on almost all projects.
When activity is been shown as a box, with arrows between the activities depicting the dependencies or precedence between the activities this is referred to as activity-on-node (AON) where a node is a point on the diagram where the various lines (dependencies) come into or come out of. Another approach called activity-on-arrow (AOA), which start at one milestone, and point to the next milestone. The visual importance is turned around in this method—milestones become the most important point and the relationships between those milestones are the activities. , while the activity-on-node technique focuses people's attention on the work to be done, activity-on-arrow focuses people's attention on where and when the various activities come together—the integration points of the schedule.
The critical chain method is an approach to project management that uses a more resource-focused than the task-centric. In
critical chain management, resources are not allowed to be partially allocated to tasks, they must be 100% allocated. Almost
all projects have some level of resource constraint, and they need to apply some discretionary logic (resource levelling) to keep within the limits of the resources that they have available. The major difference is that the critical chain method substantially reduces the amount of time available to complete each activity.
Project schedule management artefacts
The inputs to schedule management include:
* the project scope baseline. The scope baseline defines the work that needs to be planned
* enterprise environmental factors. These are likely to include specific tools and processes for developing and approving the schedule. There are many tools on the market and each one will have slightly different process requirements needed project scope statement
* organisational process assets. These are likely to include schedules used on similar past projects. Project managers should look through any previous schedules to understand what can be reused from past work, and what was successful or unsuccessful. (The post-implementation reviews and lessons learnt from previous projects should provide some of this detail)
* resource calendars
* project management plan. as the collection of sub-plans, the project management plan integrates all of the other plans, and, in so doing so, also generates risks
* work performance information
Outputs of the project schedule management include:
* activity list. A list of the various activities that need to be completed.
* activity attributes. Scope descriptions, predecessor information, successor information, input artefacts, output
artefacts, key roles, costs, calendar usage, constraints, requirements, etc.
* milestone list.
* project schedule network diagrams. Regardless of how a schedule is constructed, it cannot be constructed without logic and dependencies. It is critical that this logic is understood and examined by all of the key personnel to ensure that it accurately reflects the sequence of activities that is expected.
* activity resource requirements. A resource plan indicates when resources can be made available to complete any activity.
* resource breakdown structure. In projects where the bulk of the resources are people, an organisational breakdown structure (OBS) is important. Other resource breakdowns typically used include: system breakdown structures (which break a large system down into modules and then parts), supply lists (which break the resources down into various supply types (e.g. fabrication metals, plastics, electronics, plumbing, etc.) and resource cost codes, which allow different material types to be accounted for.
* activity duration estimates. It is critical that the activity duration estimates are recorded and kept separate from the project scheduling tool.
* project schedule. The project schedule is the output (in terms of milestones, tables, charts or graphics) of the logic and durations.
* schedule baseline. When created properly, reviewed, approved and agreed, the project schedule and its accompanying data (costs, cash flow, milestone table, etc.) are baselined. They become the basis for further reporting and the key data cannot change without all parties understanding the reason for and agreeing to the change.
* schedule data.
* work performance measurements. Work performance information is the raw data that can be collected from within the project for management reports can be developed that summarise a project's past and recent progress and forecast the future.
* organisational process updates.
* change requests. In each case, the change request will need to pass through a process of communication among key participants, an assessment of the impact of the change on the baseline and a formal approval and release of a new baseline.
* project management plan updates.
* project document updates.
Project schedule management process
The process for project schedule management comprises:
* define activities. A project manager or scheduler must have a reasonably developed plan of how to undertake a project before they can define the activities or tasks that comprise the project.
* sequence activities. There are three types of relationships between activities that are important:
1. either the activities are unrelated—in which case no sequencing is required
2. the activities are interrelated in the sense that one activity produces something needed by the other activity (a hard dependency)
3. the activities use the same resources and, therefore, have a impact on each other due to the resource constraints (a soft dependency).
* estimate activity resources. Estimation is an important 'art' of projects. It is only over time that a project manager can understand how effective a
team will be at achieving a particular activity and is this knowledge that tends to make project managers come from one field and remain working in or around that field.
* estimate activity durations. While there are many ways to estimate, the most two common approaches for bottom-up
estimation are engineering estimation (expert judgement) and parametric methods. Engineering estimation simply means the best guess of someone who has done this several times before. Parametric estimation is really comparing the work involved an activity to other work items, taking into account some parameters of the work.
Almost all estimation now is completed using the three-estimate technique. This technique was fundamental to the program evaluation and review technique (PERT) developed in the 1950s. Within that approach, estimators were asked to identify the most likely (tm), optimistic (to) and pessimistic (tp) estimate. The actual estimate that was used was the expected (te) estimate that was calculated as:
te = (to + 4tm + tp)/6
* develop schedule. Developing the schedule involves linking activities through their hard dependencies, and forming the 'logic' or 'logical sequences' of the activities.
* control schedule. Controlling the schedule involves ensuring that the project progresses as per the schedule baseline and the project plan and the commitments are achieved. The critical path is normally the project manager's area of focus. If the critical path is on schedule, then some other activities may be running behind, but (provided those activities have not used up all their float) they will not affect the end date. The first action a project manager takes to control the schedule is to ensure that there is some form of work performance information occurring regularly
Crashing a schedule is not as bad as it sounds. It means attempting to reduce the duration of a schedule by applying more resources. Crashing a schedule almost invariably involves increasing the cost of the project through this efficiency reduction. Fast tracking a project does not change the resource allocation at all (so both techniques can be used at once), but rather changes the dependencies. Fast tracking means overlapping the activities to some degree in the hope of decreasing the overall schedule. It is usually accompanied by an increase in the risk of rework though.