Your project - can you manage?

TWI Bulletin, January/February 1992

Jim Foster joined The Manufacturing Consultancy at TWI in 1986 after a number of years in industry. His past experience in manufacturing engineering has covered manufacture of computer systems, motor vehicle brakes and clutches and radio communications equipment.

Since moving to TWI he has worked on application of vision to control of TIG welding, design and construction of TWI's 10kW laser beam manipulation system and introduction of integrated manufacture to the constructional steelwork industry. Latterly his work has concentrated on factory communication systems, the problems of EMC and co-ordination of the EU 462 project.

Industry in the Western World has long accepted that important projects always overrun their time and cost budgets. It's treated as a fact of life believes Jim Foster, who outlines here the concepts which form the basis of project planning software packages, the management decision support data which these packages can provide and the benefits they offer in achieving cost and delivery targets.

Techniques have been developed for improving project performance. The value of such techniques in minimising (or maintaining) project timescales and costs has been recognised, but the effort required to apply them (manually) is often judged to outweigh their value.

Software packages, which ease the burden of project planning and provide excellent management tools, are now available. Personal computer (PC) versions of these packages are progressively finding favour as a basis for controlling large projects ( e.g. building a new factory) and managing those businesses in which the products are made to order. However, the responsibility for providing the input data on which these packages work and taking the decisions based on the results they provide ( i.e. managing), must remain with people.

Commercial wisdom indicates that attention must be given to product delivery and cost, which are as important as performance and quality.

What is a project?

Essentially a project is a requirement that must be met within fixed time and cost budgets. A project can be as simple as procuring a machine and ensuring that it is installed and commissioned without causing disruption to production. At the other extreme, a project can be as major as the Channel Tunnel construction. Irrespective of scale, all projects have the following, basic, characteristics:

• They are unique;
• They consist of a number of definable and inter-dependent tasks (or elements);
• They use a number of resources (people and equipment);
• There is a high level of flexibility in the manner in which they may be executed.

Projects go through the following phases:

1. Definition
   - agreement of overall technical, timescale and financial objectives;
2. Planning
   - determination of the detailed tasks to be performed, their costs, timescales and dependencies;
3. Development
   - evaluation of alternative approaches for achieving key tasks within the plan, culminating in acceptance by all;
4. Implementation
   - execution of the agreed (baseline) plan, monitoring performance to plan, modification of the plan in the light of achievement and taking action to forestall problems.

Many factors contribute to the total cost and timescale of a project; the opportunity to influence these is greatest during the project's definition and planning phases where, coincidentally, the expended costs are least ( Fig.1). For instance, the decision to refurbish an existing factory or relocate the business to a new site may be based on the recommendations of a single report. A second opinion, which could radically change the plan, could easily be obtained. It is likely that the considerable difference in cost between the old and the new plan will far exceed the cost of the second opinion. Decisions made during these initial phases will have major, possibly irreversible, implications for the project's cost and timescale, thus accuracy of initial information is essential.

It should also be noted that as the project progresses, opportunities to reduce its costs and timescale diminish dramatically. Consider the example of the relocated business; at the point where the site has been selected, and the building erected, the only possible changes are to the layout and manufacturing equipment - these are unlikely to have any significant effect on the project's overall cost and/or timescale.

Projects usually fail to meet plan for one of the following reasons:

• Major unforeseen events ( e.g. bankruptcy of a supplier);
• Poor project management i.e. failure to identify critical paths at an early stage, and to ensure that tasks are completed to plan and/or action is taken to overcome local problems;
• Lack of commitment to the project plan because it was poor in the first place ( i.e. it did not cover all necessary tasks and/or used unrealistic time/cost/resource requirement estimates) or because its users were not involved in its construction.

The majority of failures appear to stem from poor planning.

Planning

The '5 P' plan is the key to sound project management, namely proper planning prevents poor performance. The planning phase is crucial and consists of four essential elements:

Definition

- splitting the project into a workable number of tasks, then defining these and their logical dependencies clearly and unambiguously.

Each task must be self-contained and capable of being made the responsibility of one department (or organisation); they may, in turn, be broken into lower level tasks as subsets of the whole project. Figure 2 provides an example of top level task and dependency definition. The logic of the dependencies may be expressed in words (as Fig.2) or, in the case of computer based planning packages, may be displayed as a dependency definition diagram ( Fig.3).

Fig. 2. Definitions.

Allocation
- attaching to each task the necessary resources ( e.g. people, machines) and estimating their effort requirements ( e.g. man-hrs) to complete the task ( Fig.4). It is also necessary to assess the capacity of each resource ( e.g. only available 20 hr/wk), as this will affect the time taken to accomplish the task. This is perhaps the most risky element of planning, because tasks are unique to the project and require careful estimation. There is always concern for the accuracy of estimates; however, if the tasks are well defined, estimation of their needs is likely to be reasonably accurate. Furthermore, inaccuracies in estimates of individual tasks are likely to compensate for each other when considering the overall project estimate.

Fig. 4. Allocation

Charting
- of the data derived from task definitions and allocations, as follows:

• Gantt charts (named after their developer and often referred to as bar charts), relating tasks, resources and task durations ( Fig.5a). Charts of this nature give a visual impression of the sequence in which the tasks are to be undertaken and which resources are allocated to them.
• Resource utilisation summaries giving a summation of resources required for individual tasks ( Fig.5b). These charts provide a view of the individual resources in a way which permits the extent and timing of their under/over utilisation to be assessed. Charts of this nature are often displayed (by computer based packages) as workload histograms.

There are two distinct approaches when producing these charts, namely:

• Aiming for the shortest possible timescale in which to complete the project. This ignores all resource capacity constraints but does indicate the resource overloads - which the Project Manager must take action to alleviate ( e.g. draft in more people, work overtime, etc). Figure 5 illustrates this situation.

  or

• Recognising the resource capacity constraints and ensuring that they are not exceeded. This frequently results in increased project timescale. Figure 6 illustrates this approach.

Critical Path Analysis (CPA)
- of the whole project based on the task dependencies and allocations. This may be documented as:

• Activity on arrow - the traditional, manual approach, illustrated in Fig.7a; or
• Activity on node - sometimes referred to as precedence and easier for computer based systems to accept, illustrated in Fig.7b.

Whichever documentation technique is used, the object is the same, namely:

• To establish those tasks for which any change in duration will affect the project end date. These are the critical tasks and they are linked by the critical path.
• To establish the amount of time by which other (non-critical) tasks may exceed their estimated duration, before they also start to affect the end date; this is termed float.
• To generate a snapshot of the project, showing those tasks which must not be allowed to slip and the scope which exists, elsewhere in the project, for ensuring this by moving resources from non-critical tasks.

Critical tasks/paths and the float in non-critical tasks may be evaluated manually ( Fig.8). The earliest start date is based on starting all possible tasks at day 0, whilst the latest finish date is based on maintaining the earliest finish of the project's last task. Manual evaluation of critical paths is laborious for all but the simplest of projects, computer tools readily perform this task and generate the desired network diagram ( Fig.9).

Fig. 8. Manual evaluation of the critical path

Implementation (of plans)

The project plan, once agreed and accepted by all parties concerned, forms the project's baseline; from this the final (implementation) phase must be controlled.

All future actions must be monitored against this baseline and the working plans revised whenever necessary to accommodate the current situation. Monitoring of current situation versus plan (tracking) is vital; from this stems early warning of impending problems. Given early warning, it is usually possible to initiate actions ( e.g. re-schedule future tasks, re-allocate resources) which will alleviate the worst of the problem. Using purely manual techniques, tracking requires considerable effort.

Before the advent of computer based project planning/management tools the enormity of updating plans frequently caused the technique to be badly used ( i.e. not updated, therefore carrying untrustworthy data) or ignored. Frequently the plan was of little value and simply became wallpaper for the project manager's office! However, PC-based project planning tools ( e.g. PMW* as used by TWI's Manufacturing Consultancy) have changed this; there is no longer any reason to believe that the value of project planning will be outweighed by the cost of maintaining the plan.

Project planning software tools are capable of rapidly generating updated plans for single or multiple projects (which share common resources), being networked and importing/exporting project data from/to other software tools ( e.g. staff time bookings from/to spreadsheets or databases). Such tools are finding favour in the operation of complete businesses as well as for managing individual projects.

*PMW (Project Manager Workbench) is a product of Haskyns Group plc

Why bother?

The benefits stemming from sound planning and management of any project maybe summarised as:

• Increased confidence in the predicted cost and timescale of the project through systematic establishment of the tasks to be undertaken, realistic assessment of resource requirements and application of understandable/proven planning techniques.
• Improved control of the project arising from confidence in the plan, visibility of progress and the ability to test 'what if?' scenarios quickly.
• Team focus arising from the close and early involvement of all team members when defining tasks and allocating resources. This focus is further enhanced by the provision of clear timescale objectives and full project knowledge (through the plans) for each team member.
• Reduced cost by ensuring that the project meets time and cost budgets, thus avoiding overrun/overspend which are the hallmarks of lack of project management.
• Improved business performance by removal of the knock-on effects of late projects.

To gain these benefits it is necessary to plan properly using established techniques, and to manage effectively, aided by modern software tools where appropriate. Reasons for failure to plan and manage projects appear to stem from lack of:

• A perceived need for project management;
• Trained staff and/or facilities (hard/software);
• Confidence in and/or experience of the techniques and facilities;
• Experience in recognising, defining and estimating all project tasks;
• Expertise in recognising and acting upon early warnings of impending problems.

TWI has recognised the value of software project planning packages and uses these where appropriate. TWI also recognises that many of its small to medium members may not be able to achieve sound planning and management of key projects for the reasons given above. The staff of The Manufacturing Consultancy (TMC), who have considerable experience in successfully managing both research and manufacturing projects up to £20M in value, provide a resource which members can use to eliminate these problems. The combination of experience in the field, the skill and tools with which to analyse problems, the freedom to develop unbiased solutions and the tools with which to do the job efficiently and make supported decisions, are the ingredients of a full project management service.