Planning is an unnatural process. It is much more satisfying to do something. The nice thing of not planning is that failure comes as a surprise rather than a long term worry and depression.
Initial Planning
A project may begin in many ways: a flush of inspiration, a response to need by a customer or client, an exploration of an issue or personal interest, to it can be part of a larger programme.
Having outlined the product requirements, measures and benefits the next step would be to come up with an initial plan. This should indicate the outcome of the project and the process by which it will be delivered. The outline plan needs to provide the means to decide whether any further work should be carried out.
Projects can be characterised by order and chaos. Chaos comes first, followed by well-developed system. There is an input and output to each project.
Managing the project scope is important. Scope creep exists when a purpose of a project is change on many occasions until it does not longer resembles the original concept. Another important point is to ensure that all parries agree in what the project is going to do. Was generally includes the scope plan to generate the process scope and product breakdown to generate the project statement. And again, it is very important to obtain acceptance. This may be via formal sign-off process where everyone agrees to the scope.
There must be well defined route for the planning to take place. This is not always very clear and a generic model of planning is very difficult to construct. Planning as a process involves consumption of resources and the costs. The project manager must balance the resources and costs. We need to avoid the costs of the chaos that would otherwise take place with an unplanned activity. We must also provide the basis for formalised evaluation process and identifying problems in advance.
Investment at the planning stage generally brings significant benefits. Plans are used to help decision-making and guide future activities. A well-balanced plan will guide the actions of the project team without the need to define the absolute detail what each person will do for every minute of every day.
Project plans should also change as circumstances change. Changes can be made with a new actions. Projects need to gather data from relevant sources, therefore project manager needs both, collect the data and the information.
In a plan, we generally define what and how tasks will happen.
The inputs are the basis on what is going to be converted by an activity. The outputs is the project plan, or more specifically, the project proposal. There should be a circle of many reviews and suggestions before the final plan is produced. Once the project team is happy with the formal plan, small adjustments can be made.
In planning, sometimes being as precise as possible may be less beneficial than being as accurate as possible. There is clearly a role of detailed planning, but not before an overview plan has been worked through. Evaluating the overview reviews some fundamental flaws in assumptions.
The revision process considers the necessary subproject, the result of any numerical analysis, the element of "gut feel" as well as experience.
As complexity in projects increases, so does the necessity for a formalised plan. Traceability has become a major issues for many companies, and the same is required for a project plan. In an event of undesired result, the project manager can show that they took every possible precaution to ensure that the result was positive.
Good planning allows breaking large tasks into manageable activities, determining logical sequences of activities, providing a logical basis to making decisions, providing the framework for the assessment and essential for the revision process.
The PID
Often PID is the output of the overview of the planning process.
PID includes:
The project brief - a statement of the requirements of the project as understood at this stage of the process
Project team management structure - who is to be involved and the governance
Job descriptions - The roles that each team member will take
The project approach - outline description on how the requirements in the brief will be delivered
The project quality plan - outline basics steps to ensure that the requirements are delivered including checks on the process
The project plan - a basic plan showing major activities to be undertaken, their sequence and duration
The business case - the first attempt at evaluating the cost and benefits of a project
A risk log - a list of major risks identified for the project and any actions taken to minimise them
Project controls - at the most basic level, how will you know if the project is going to deliver what is has promised, at the time required and the cost stated, while maintaining all the essentials
A communication plan - a statement on what will need to be communicated to which stakeholders in what format and when
The PID is one to capture the requirements of a project. The level of formalisation differs between the organisations and often between projects in those organisations. Generally the higher level of complexity will require higher level of formalisation. It will highly depend who are the users, decision makers and owners.
Time Planning
The initial plan was now created and approved. Now it is time for detailed project plan. This is not a trivial process. We need to plan for most possible scenarios as well as eventualities. It is all to answer the question "what happens if".
The most basic approach is to deconstruct the project intro into its constituent products and parts. These than need to be compiled into a product time plan and feed the basis analysis and communication. This is when the Gantt chart tools can be very helpful.
Determining which parts to include into a plan and how to make them work together is where the 'art' rather than the "science" of managing projects is needed.
The production of plans is often treated as an end in itself. The planner must have the opportunity to analyse the system, to review opportunities for improvement and problem prevention and setting out the basis on which the project will proceed.
Having an overview model it is now time to put place into detail. Graphical techniques are preferred. The methods have been formalised for over 50 years.
Work Breakdown Structure
With some small projects many people are able to break this up into list of things to do in their heads. But there are downsides of this approach: vulnerability of the project to that one person and inability of the plan to be analysed and communicated. In order to be more systematic about it we generally use WBS.
The breaking down of large activities into manageable units. This chunking gives people responsibility for a manageable part of the project. It also facilitates the financial control and tracking of resources.
The breakdowns can be by activity or even by functions or physical groups.
For a project manager to perform the work breakdown is not a trivial task. We must remember, that the project is not about slicing it up, but the opposite taking the various parts and making them work together.
Time Plan
Estimating is a key part of project planning. These are generally "guesses" therefore, we need to be careful how we treat the numbers from this process and not do it in a hurry and making rough estimates.
The estimation process will vary from collecting estimates from other people in preparation of the proposal for cost-benefit analysis. Estimation should also continue during the project life cycle and the accuracy of estimates is going to get better.
"Activity on node" and "Critical path analysis"
Using the "Activity on node" technique, the construction of the project plan combines the Work Breakdown Structure to identify all the activities, the sequence to which they need to be performed, and the estimated duration of each activity.
If the logic of the project can be identified, the project manager can develop a plan to represent it, allowing better analysis of the activities. The visual clarity of activities as boxes and linking them together is appealing. However, for any two activities, there are four ways in which they can link.
Finish to start - the second activity cannot start until the first one is finished
Start to start - the second activity cannot to start until the first one has started
Finish to finish - the second activity cannot finish until the first has finished
Start to finish - the second activity cannot finish until the first one has started
We can amend the activity boxes to add more information.
- EST - Earliest Start Time
- LST - Latest Start Time
Total float - the difference between LST and EST
We can use this to determine how long the project is going to take. The path through the activities which take the longest is know as the "critical path".
We firstly determine the overall project duration.
In the diagram, these are the three paths between the start and the finish A-B-E-H , A-C-F-H, and A-D-G-H. This first stage of the analysis is known as forward pass. The subsequent calculation is the reverse pass to determine the critical path through the activities.
Forward pass
The forward pass determines the EST and start at the left-hand-side of the diagram with the EST of the first activity. The next activity, A, cannot start until this has finished and so the EST for A is the EST for the start plus the start duration (point 0). Activity A has a duration of 5 days and so the EST for B,C,D are all 0+5 = 5
Continuing with the EST:
EST E = EST B + Duration of B = 5+ 3 = 8
EST F = EST C + Duration of C = 5+4 = 9
EST G = EST D + Duration of D = 5 + 5 = 10
They are now entered onto the diagram. Activity H presents a small challenge as there are 3 possible ESTs:
EST E + Duration of E = 8+6 = 14
EST F + Duration of F = 9+7 = 16
EST G + Duration of G = 10+5 = 15
Activity H cannot start until all 3 reseeding activities have been completed - in this case at time 16.
Following this through to the end of the network gives the EST (FINISH) as 16+8 = 24.
The reverse pass
The reverse pass starts from the end of the network and assigns the LSTs for each activity. The Last at the finish is assumed to be the same as the Last for that activity with the meaning that we want the project to be completed as soon as possible. Working backwards the LST for H is the LST for the Finish minus the duration of H. That is it is the latest time that H could start without delaying the entire project. This is 24-8 = 16 the same as the EST. Continuing with the analysis, the LST for each of the activities E,F and G can now be calculated:
LST E = LST H - duration of E = 16 - 6 = 10
LST F = LST H - duration of F = 16-7 = 9
LST G - LST H - duration of G - 16 - 5 = 11
and similarity for B,C and D
LST B = LST E - duration of B = 10 - 3 = 7
LST C = LST F - duration of C = 9 - 4 = 5
LST D = LST G - duration of D = 11-5 = 6
The LST for A is another case, where there is more than one possibility. The three possible LSTs are:
LST B - duration of A = 7 - 5 = 2
LST C - duration of A = 5 - 5 = 0
LST D - duration of A - 6 - 5 = 1
The choice is the easiest of the LSTs - 0. Checking back to the start activity confirms that this is correct.
The additional task with the Activity-on-Node technique is to calculate the 'float" in each activity.
Float = LST - EST
This is than inserted in the bottom right hand side of the activity box. When there is no float, it means that the activity is critical and any delay in this activity will delay the project as a whole.
The critical path through the network is therefore
A-C-F-H
Knowing the critical path has significant implications for the project managers. In the example above we know that we can tolerate no delay without impacting the end date. This means that the project manager should focus more attention on activities A,C,F and H. In this case the overall duration increases by one day as well, so it is vital that the project manager does not focus all of his attention on the critical path and forget how all the other tasks are progressing.
Microsoft project, uses a slightly different format of A-o-N diagrams as it takes absolute dates rather than relative time for EST and LST.
The activity description is given at the top of the box, the next row down contains activity number and the activity duration. The bottom row of the box contains the start and the finish times of that activity. For example, the duration of 7 days does not include weekend working, hence the start on the Wednesday and not finish until Thursday the following week. With Microsoft Project the calculation is done for us already. However, it doesn't allow us to see whether there is any float in the activity. But this could later be covered in Gantt charts.
Scheduling
Knowing the sequence and duration of activities is a major step. But we need to be realistic with out plans: people are not available when needed, resources have long lead-times, there could be clashes with other project schedules. Therefore the activities will often need to be changed.
As per the diagram there are relevant inputs of the scheduling process. The network provides the logic in which activities need to be carried out. Having completed this, the input of the time estimates might show that the time project requires for completion is longer than is available. This might require a reconsideration of the network and the time estimates.
Project manager also need to allocate resources and need to have the knowledge of capabilities and limitations of the resources. We also can take advantage of the resource calendar. Machines might be able to be utilised 24/7 but not humans. The calendar are those times when resources are available allowing for weekends, bank holidays, personal holidays and sick leave. Resources availability must be taken into consideration and projects rescheduled.
Gannt Charts
Te purpose of Gantt charts is to illustrate relationships between the activities and times.
In the example activity A is represented by a shaded bar starting at time 1 and finishing at time 3. Multiple activities can be built up on the same chart. In the example the WBS identifies the activities and the time constraints - the start and end dates.
When moving from WBS to Gantt chart there is a level of logic established, time goes from left to right, activities are arranged top to bottom. The tasks have the rough times allocated to see whether they meet the start and end times.
A great ability of the Gantt chart is that it can communicate the plan. They have limitations however, when the tasks use the same resources. We should not, however, perceive the charts as a statement of reality.
Tools, such as MS Projects highly influenced organisational practices recently. This also is being replaced by PRONCE 2 2009 trining for employees.
Additionally, many really excellent organisations do not use Critical Path analysis in their practices. Whiteboards and sticky notes are still being utilised.
Re-thinking time planning
Applications of Theory of constraints provided significant benefit for repetitive operation and well as project management. This is the Critical Chain Approach of project management.
Projects that run late, over budget or fail cause considerable problems for businesses. The problems that we see regularly happening is Uncertainly.
This may be related to the process or the outcomes. Unfortunately in some large projects, many projects are incapable of handling change. Additionally, the system does not have credibility with the uses and those who perform the tasks. Under pressure, people focus on the short term objectives of completing activities. The longer the project duration, the greatest changes of changes.
We must understand that with the planning and CPA as well as PERt the goals are based on estimates, which in itself contains uncertainties. Activities will sometimes run on time, often late, but almost never early. Even if we apply safety margins based on our experience it does not assure completion on time.
Multi-tasking
We will present the concept of multi-tasking in the following example. Imagine that we have 3 projects to work on: A,B and C. Each of them is going to take 10 days for us to do our part of the work.
If we do them in sequence, the lead-time (start to finish) will be shown as above. This is simple enough. However, if each of the projects are being broken into smaller units. The effect of breaking the project and doing half of each then returning to finish the reminder.
The effect is that the projects will take considerably longer. I a case of one new product development team, we can only imagine the effects of having an average of 12 projects per person. We generally, make assumption, that we can leave and pick another project without any loss of time. In some software development companies, developers would generally loose 1-2 days by switching between tasks. They needed to refamiliarising themselves with the logic of the task and the management system should have identified this and work to prevent such waste of time.
There are, of course, projects where they are delays of getting information or results and where some multi-tasking is inevitable. For example, enlightening industry fluorescent lamps guarantee around 10000 hours of operation. In one year, they are 8760 hours and it is no realistic to perform such live tests. It would be pointless waiting for test results to come through before starting on other work and the development team needs to move on to other project.
Therefore, projects require planning taking into account human behaviour when estimating.
Typical problems with estimates include:
- we use estimates inappropriately - people being asked on rough estimates how long a particular activity will take only to find out that it becomes the target time,
- we use inappropriate data to built estimates - people either take unrepresentative previous experience or not checking whether it was a fact or good representation of the reality
- the estimates are used out of context - estimates are guess
The estimating
we can view the construction of estimates as per below picture.
For example, we are asked on how long it will take to prepare 2000-words report on a subject which we are familiar with. We know from previous experience that it will take approximately 2 days. It would be dangerous, however to promise that. To ensure that you will finish on time, we would add to the task time. There may be other things you need to do that you have already planned for, another day for interruptions, and another half-day for any problems - the computer crashing, for example. Our 2-day activity now takes 4.5 days, or even 5 days, just to be sure. It will almost certainly take five days, or usually, even more.
Human nature for any of us is also to leave the project until the last possible minute. This is called student syndrome.
The result of this is that time might of been wasted on the project. This is typical.
Managing by constraints
The theory of constraints (TOC) was the result of application of structured logic approach to problems in manufacturing environment. It targeted the way in which production lines has being scheduled and the flow of goods was managed. It related to managing systems by focusing on constraint - often called a bottleneck. In a production system, a bottleneck can usually be identified, by the pile of inventory waiting to be processed by the part of he process. It determines the ability of the production system to do work and earn revenue.
The stages of TOC approach are as below:
- identify the constraint - the critical path and the critical resources
- exploit the system constraint
- subordinate everything else to the constraint
- evaluate the constraint
- go back and find new constraints, repeating the process
In the production situation, identifying the constraint provides the immediate benefit. Let's see how the method can be applied in project management.
Step 1. Identify the constraint
In projects, the constraint is anything that determines the project duration, in order words, what is stopping the project from running faster. This could be:
- the critical path of the project,
- the availability of resources
- dates that are fixed into a schedule and cannot be moved
- behaviours, policies etc
The constraints will be different things in different time during the project.
Critical path is only a starting point in calculating the project duration. Considering the constraints. Removing resources also cause delays as projects are interdependent within the activities. This is the reality of modern project environment. After determining the constraint of resource availability, we need to consider the fixed dates in the schedule. For example when some items need to be signed off. If board meetings are only every 2 months, the projects might be suspended for 2 months. Many projects run entirely on scheduled dates for every activity fixed in advance. This way every activity becomes a constraint. The last of the constraints are those imposed by the organisation or individuals. Some business rules may cost delays, for example meetings a certain day etc.
these constraints are generally not very obvious and they will change throughout the entire projects. Projects generally take significantly longer than they need, because of failure to consider the impact of constraints.
Step 2. Exploit the constraint
The most important thing in a project plan is the constraint. Having identified these, anything that stops the project from completing on time needs to be removed.
Step 3. Subordinate everything to the constraint
Even if we keep all the tasks of the project at 98% completion on time, the 2% can delay the project. This is because the 2% are critical tasks. That's where the subordination come to play. Make the constraint the point around which schedules are based and ignore local efficiencies that do not consider the system as a whole. Activity completion should not be the measure - project completion should be. Where the constraint is a resource, it needs to be kept as the most important thing. If the constraints are the people, because most work needs to be undertaken by one person. Therefore the work must be ready for him to complete his tasks. People should also not be interrupted until the tasks were complete.
Step 4. Elevate the constraint
Elevating the constraint means increasing the flow, through that part of the system - removing it as a constraint. Either it is a machine or an individual it needs to have its capabilities increased if the system is required to work any faster - either through working longer or through adding extra help (another machine or person). Sometimes another machine is required stand-by, just in case a machine break down, this will elevate the constraint.
Step 5. Go back to Step 1
In a repetitive operation, the last thin is to go back and find next system constraint. In projects it happends as a matter of routing anyway. The project manager will need to check progress and recognise where constraints change. The resources and activities will need to adapt accordingly.
Critical Chain approach
In projects the true consequences of not finishing should be carefully evaluated in financial terms. Determining these costs provides the justification for the additional work to be followed in the planning stage.
Estimating
First, the principles.
- Estimates (both initial and to-finish) should be based on the activity times only, with no safety added. This relies on a number of changes including buy-in from all people estimating and the commitment to acceptance that 50% of activities will finish early and 50% will finish late. Finishing late is therefore expected for half of the activities.
- Safety should be included at the end of critical path - not before. Where there are feeder paths (parallel activities that lead to the critical path), the safety buffer should be placed at the point, where the feeder joints the critical path.
- Time plans establish precedence relationships but should be treated as overviews only. The nature of project management needs to reflect the dynamic of the actual situation and accommodate changes as they occur. Giving this uncertainly, the parties involved need to be given a regular updates on when their input is required. It reflects the need for greater communication to ensure critical activities can proceed without delay.
- Progress should be monitored by the critical path, rather than % complete.
The constraint might be outside the firm.Sometimes we are waiting for other external information. We should identify this constraint and buffer it accordingly.
One of the major challenges for project managers is keeping up to date the plans. They may re-plan the project on a daily basis and re-issue copies to everyone in the project whenever they are changes. This can be very unsettling for everyone concerned, but critical chain plans with buffer and due-date constraints removed are far more stable. This stability is a main benefit of critical chain benefit.
Additionally, when non-critical activities become critical, it is only necessary to consider the criticality. Non-critical activities generally become critical due to resources contentions. These resources ( a person, a department, external organisation) are now the constraint and should be protected by a buffer (some time-slack in front of them). The constraints should be managed in the same 5 Step process.
This added complication to the critical path represents a formation of compound series of activities which has been termed as critical chain. The TOC approach results in new approach to performance, particularly at the local activity level. This alone is considerable potential benefit.
Control
Project manager has the ultimate responsibility for the control - by determining the issues of importance, their measures and then devising the system by which they will be monitored and corrective actions implemented where necessary.
The break with due dates has been identified as one of the features of critical chain methods. Instead of waiting for the due date to arrive before starting the activity, we should continue directly to the completion of a subsequent activity.
Where due dates are used, there is neither a requirement or incentive for an early finish. Under critical chain, this must be used, and any early finish added to the buffers. The role of the project manager therefore changes. Far from watching the dates, the issue becomes managing the hand-overs between the activities, ensuring that early finish is encouraged and that subsequent activities are ready to start. This is the major information handling and communication role. Managers of projects undertake this anyway.
Comments