As it may be noticed, the job is not as easy as just scheduling the activity by determining the time it will take to accomplish it. It does require a great deal of additional planning and scheduling if we are to attain the desired objective. The figure next page shows how resources should be loaded to an activity in the master schedule.

Therefore, the question know is to decide how much information should be enough input for the schedule and how it should be handle to avoid confusion while at the same time certifying that the schedule may go as planned by communicating all requirements to all interested parties within the project team.

The answer to the first question is an easy one, all information related to the activity must be made available to the schedule. As for the second question, however, it would be very confusing to present everyone within the project team with information other than that concerning their responsibilities. The best solution is to produce additional schedules out of the master one to communicate with the corresponding trades and trigger action from them when required.

The Original Schedule Analysis

Once all activities in the original approved schedule have been loaded with resources and responsibilities as explained before, the original schedule analysis can start.

This task is mostly performed by charting resources against time during the project duration as shown in the figure next page.

These charts will help the planners/schedulers in their job to create resource pools with enough anticipation to avoid production bottlenecks and low productivity areas.

Cash flow analysis for the project is one of the immediate results of this analysis. Regardless of the procedure used for time distribution of the activity cost the planners/schedulers can calculate the amount of cash needed for each project period just by adding the expenditures of the activities following in that period.

Planning a company's cashflow is an important part of good financial management and its purpose is to identify cash shortages or surpluses and to deal with them in the most efficient manner.

Project management must supply the company with all required information about the projects needs with enough anticipation to allow company's cashflow planning accordingly.


The concept of cashflow is not an obscure one. Either the company has a certain amount of cash or it has not. And a lack of cash is critical. A company can sustain losses for a time without suffering permanent damage, but a company that has no cashflow is insolvent and in imminent danger of bankruptcy, no matter what profit picture the future may be showing.

Projects may be significantly affected if cashflow planning is not given the necessary attention.

From the project management point of view there are three major concerns when dealing with cashflow planning:

1.- Project cost must be scheduled according with company's cashflow capabilities.

The project financial representative must establish his cashflow needs according to the Projects consolidated budget and schedule. The company's finance department will come back with the financial constraints ruled by the company's cashflow.

Company's priorities will play an importantrole in deciding what project must be developed first.

2.- Approved projects' cash flow variances may affect the company's overall cashflow by the lack of return in cash not expended or by the cost of meeting unplanned cash requirements.

3.- Projects cashflow analysis is the foundation for capital investment appraisal and management decision-making.

The standard method of evaluating potential investments is the cash flow analysis, which provides the timing and the amount of all projected capital and operating expenditures and related revenue. Annual net cashflow equals revenues minus expenditures calculated for each year of the economic life of the project.

The reliability of the projects evaluation depends on the accuracy and completeness of the cashflow analysis, so it is important to include all related costs, including direct support costs. Economic evaluations of investments involve comparison of some alternatives including taking no action.

If there are several projects needing control by the same planners/schedulers, they will be able to establish the cashflow needs for the group of projects by just adding the individual cashflow need for each project.

The lack of adequate tools, techniques, and system design knowledge has been primarily responsible for the historical difficulties with project management. Most of the traditional scheduling techniques are based on the Gantt or barchart, a tool which has been in common use for over 50 years. Although it is still a valuable tool, its use is limited in the scheduling of large scale operations.

In particular, the bar chart fails to delineate the complex interactions and precedence relationships existing among the project activities. In addition, it does not lend itself to mechanization through the use of a high-speed electronic computer and thus cannot utilize many of the scientific management techniques that computers make feasible.


The Network-Based Approach

In recent years, the problems of project management have received concentrated attention. The work of several independent investigators led to a family of techniques that represents a major breakthrough in project planning and scheduling techniques. The complexity of the systems required for military projects has given great impetus to this development.

The Polaris missile program led to the development of PERT, Program Evaluation and Review Technique by Lockheed Aircraft Corporation along with the US navy special projects office and a consulting firm called Booz, Allen and Hamilton. PERT is a probabilistic approach specifically designed for new projects or projects involving a high degree of uncertainty.

However, the normal activities of industry also involve work which fits into the project concept quite naturally. DuPont and Remington Rand Univac interest in optimization of such projects as routine plant overhaul, maintenance, and construction work led to the design of CPM, the Critical Path Method in the late fifties.

CPM is a deterministic approach requiring not only a high level of professional discipline but a complete commitment of top management to become a useful management tool.

Known by these and hundreds of other acronyms, these techniques share a large number of common elements. Chief among these is the use of a network flow diagram as a model of the project's precedence relationships.

The network represents all the activity paths or sequences that must be accomplished before the project's objective can be achieved. The longest of these sequences is the "critical path," the identification of which permits management to focus its attention on the progress-pacing activities of the project.

Much of the early success of PERT/CPM was based on the explicitness of the project plan, this explicitness being essential to the construction of a network.

Being explicit about what was to take place at some much later time was a new experience for many.

Improved communications among those concerned with a given project was the result most frequently cited by the individuals involved in the early attempts at networking the project plan.


The Role of Networks in the Engineering Management System

The project management system requires the existence of means for the description and evaluation of alternative project plans. The network models described here are means of depicting a particular project plan in such a manner that evaluation is not only possible but is in fact a logical extension of the model.

A given model of this basic type will describe one alternative. Other models will be required if other alternatives are to be examined.

The modeling process will be described on the assumption that the project management consultant is preparing the network based on information he extracts from the project manager and others selected as sources of information.

It will further be assumed that the project management consultant knows little of the technical details of the project and the project managers know little of the network concept being used.

Network Construction

The network concept involves the graphic representation of activities and their precedence requirements. Activities are elements of the project representing logical subdivisions of the work to be done. If you considered preparing breakfast as a project, pouring a cup of coffee could be an activity.

The level of detail used depends upon the degree of control desired.

Precedence requirements indicate which activities must be completed before a given activity can proceed.

The project network may be formed in several ways. One of these is to start at the realization of the end product or objective and work backwards in time in a step-by-step fashion, determining what work must be completed in order to start a given activity. Another approach is to list randomly all the jobs having a bearing on the project and to determine their technological relationships as the diagram is developed.

We want to be able to identify each activity by a work item number for use in a computer analysis of the network. Activities may be numbered in any fashion, as long as a number is not repeated, without creating any logic or identification problems.

Computer processing of networks depends upon such numbering, but most computer programs have the capability to take any activity numbering and renumber with predecessor less than successor for internal usage, reverting to original numbers in the output.


The belated discovery of activities is an inherent property of the network design of a project plan. The fact that such activities may cause substantial redesign of the network not only slows down the networking, but also tends to inhibit the project planners in their search for activities overlooked previously. An important psychological factor in the planning stage is inertia.

The networking system should be capable of maintaining an up-to-the minute graphical representation of the project plan as it is being formulated. PDM precedence diagramming system is quite effective in this respect when compared to the ADM arrow diagramming approach. Planners sitting around waiting for the networker to incorporate their latest disclosure into the ADM network tend to lose the enthusiasm with which they may have started.

LAG Factors

There are situations in which establishment of the proper precedence relationship gives an erroneous representation of the project.

Networks Level of Detail

A general guideline for determining the level of detail is that it must be keyed to the level of management using the network. Hierarchies of networks will be employed in more complex, large scale projects. A major aircraft design project, for example, might have three levels of networks.

The top-level network might summarize the ten major subdivisions of the project into 300 activities. The next level might involve describing each of the ten major project subdivisions with a network of 1,000 activities for a total of approximately 10,000 activities used in this intermediate level.

Each intermediate network could break down into ten l,000-activity networks employed at the actual operating level, resulting in a total of 100,000 activities being used at this most detailed level.

At any given level of management, there is always the question of how much detail is appropriate. Some organizations attempt to control level of detail by specifying that the network shall consist of a total number of activities ranging between X and Y.

This approach applied to projects of varying complexity can produce networks with widely varying levels of detail. A more sound approach to obtaining a consistent level of detail involves specifying that no more than X percent of the activities shall have a duration of more than Y or less then Z time periods.

Some more specific approaches to making level-of-detail decisions are:

1. For the resource entities being controlled by this management level, one activity should end and another start at the point where a different resource comes into use on a given work item. This might be the basis for deciding to use the three activities, Layout Site, Clear Site, and Grade Site rather than the single activity, Prepare Site.

2. If some other activity is dependent on the first portion of the activity in question but is not dependent on its full completion and the overlap amounts to as much as one of the time periods used in duration assignments, the activity in question should be divided into two activities at the point of dependence.

3. If there is considerable doubt left as to whether or not an activity should be broken down into more detail, one should remember that the network analyst can later combine activities without the project manager's assistance, but will require assistance to further subdivide activities.

Combining Subnetworks

No matter how involved and complex the project, the rules for combining activities and precedence information into a network still hold. Often with very large projects, whole subprojects are performed independently but simultaneously as a part of the overall project. It is then possible to network the subprojects to reduce the complexity and then combine these sub-networks. Contractors or managers in charge of the subprojects may then have their responsibilities scheduled separately.

Overall management's total network can correspond precisely to the sub-networks without requiring an overall network to be constructed separately. Under these circumstances, responsibilities are split out before the network is constructed instead of after, and the complexity of the network generation is reduced rather than increased.

Any interface activities are identified with unique symbols to mark the points where the networks rejoin. These activities on the overall network contain the corresponding activity numbers from all subnetworks.

CPM Implementation Recommendations

The project management consultant must acquaint himself with the project objective before starting the construction of the network model. He may do this in any or all of the following ways:

1. Reading the specifications
2. Revising drawings
3. Looking at artists' concepts of finished product
4. Discussing the function of the end product
5. Seeing a similar end product in use.

Then, he determines the management objectives, discussing with the project manager the question of which of the following objectives should govern the project management system operation:

1. To meet a deadline for project completion
2. To minimize project length
3. To minimize the sum of direct and indirect costs
4. To minimize the variation in resource requirements
5. To limit the maximum resource requirements of one or more resource types
6. To communicate project information to various levels of individuals concerned
7. To co-ordinate the project with interface groups

The consultant is now ready to undertake the networking of one of the alternative approaches to the project. Subdividing the project as discussed earlier is the first step which should be attempted.

After either arriving at subdivisions or determining that no reasonable basis for subdivision exists, the planner should obtain a summary narrative from the project manager describing the approach to this portion of the work. This narrative should be in the order of 5-15 minute discussion of the intended project plan.

The consultant is now ready to start networking. It is possible to proceed by eliciting responses from the project manager to the question, "What other activity must take place?", and putting the response on paper as an activity node.

In general, one should put activity descriptions, abbreviated if necessary, in the nodes rather than code letters or numbers referring to some separate list. The network should be a document that can be read directly. Reference to a coded list destroys the flow of logic that one should perceive when examining a network.

Precedence can be ignored at this time and the result will be a sprinkling of unconnected activities over the worksheet. Some activities are obviously near the start of the project and will likely be placed to the left of the sheet, just as activities likely to have few successors will be placed toward the right. It is not necessary, however, that any location rules be followed. Worksheets containing pre-printed or predrawn nodes will speed up the network building process.

There is reluctance on the part of many individuals to commit to even the most preliminary network. It must be emphasized that this first network will be re-drawn and eventually change forms and that the original network will be discarded. Even then, however, some project managers respond more readily if their tentative descriptions are placed in a list first, then put into the network format.

The network planner may well invest some time in listing activities purely for the purpose of getting the modeling process underway. The network planner should never attempt to indicate predecessors on the list, however. Precedence information is much more easily obtained by directly incorporating each bit of information into the network as it drawn.

When the process of discovering activities begins to lag, it is desirable to turn to precedence identification, rather than try to identify 100% of the activities on the first pass. Precedence may be determined by asking this question with reference to the predecessor: "What activities cannot start until this activity is completed?" In the process of establishing precedence connections, additional activities will be discovered and redefinition of existing activities will become necessary.

This process should be continued until the project manager is unable to readily find additional precedence relationships. There will usually be some precedence relationships that are not found in this first phase, but these remaining ones can best be discovered during the duration/resource assignment phase.


Network planning differs from bar charting in that no time/resource concerns restrict the construction of a network. This time-free model tends to encourage the consideration of more alternatives than does the bar chart. When the planners begin to have difficulty in thinking of additional activities to place on the network, however, it is then time to consider activity durations and resource requirements.

As the activities are reviewed for time/resource requirements, additional activities and precedence relationships will be discovered and some existing ones changed.

At this point it will be helpful to have a draftsman or clerical assistant redraw the network, attempting to have predecessors to the left of successors and attempting to minimize the crossing of arrows.

It is impossible to avoid having some arrows crossing others, but a sense of flow of the work is best attained when these are minimized. The revised network is then ready for use in estimating time and resource requirements of each activity.

For some individuals and groups, the process of obtaining this basic network model will give new understanding of the nature of the project. In fact, an organization may find it desirable to go no further on its first attempt at network project management.

Such an approach might involve posting the network where the project manager can refer to it during meetings and shading in each arrow as the activity is completed.


ATTACHMENT I

Scheduling Expressions

Activity: Work item represented by an arrow when using arrow diagramming or by a box when utilizing precedence diagramming.

Actual completion date: The date an activity is actually completed.

Activity duration: Time allocated to complete an activity. OD stands for original duration and RD stands for remaining duration.

Arrow diagram: A logical plan to accomplish a job where all the activities required to complete the job are represented by arrows organized and interconnecting in sequential order of execution.

As of date: The calendar date equivalent of time-now for statusing purposes.

Barchart: Graphic representation of items of work displaying duration of item as a bar whose length is proportional to the time allocated to perform the item.

Calendar days: Scheduling time units for measuring activity durations which includes weekends and holidays.

Concurrent Activities: Activities which can be performed at the same time


Constraint: Any factor which exerts a time or sequence effect upon the plan activities.

Critical Path: The shortest time in which a plan can be performed.

Cut-off date: The assigned date on which accumulation of data for a reporting period
must be completed.

Data base: A collection of records for the project which allows management's information requirements to be drawn from it.

Direct cost: The portion of total cost which takes into account only material, labor and equipment assigned to a project or activity.

Dummy arrow: A dotted arrow used as a constraint in a network to display where a relation exists between tasks but no activity arrow connects them.

Early finish (EF): Computed value for the earliest possible time at which an activity can be completed.

Early start (ES): Computed value for the earliest possible time at which an activity can begin.

Event: The starting or ending point of the activities.

Float: The difference between the computed time available in which an activity may be completed and the estimated duration time previously assigned to the task.

Fragnet: Any area of a network that has been expanded into more detail and definition for analysis.

Forward pass: Procedure to calculate early event times for all activities in a network.

Gantt chart: Synonym of barchart

Hammock activity: Summary of a entire subset of activities utilized to reduce the number of project elements to track or for analysis of a portion of the plan.

I-node: The number assigned to the beginning event of an activity arrow.

J-node: The number assigned to the ending event of an activity arrow.

Lag factor: A ratio or quantity used in precedence diagramming (PDM) which numerically indicates when a dependent activity can start relative to the
duration of the predecessor.

Lag time: A time delay required by a sequential connection in a network.

Late finish (LF): Computed value for the latest possible time an activity should be completed according to CPM calculations.

Late start (LS): Computed value for the latest possible time an activity should begin
according to CPM calculations.

Lead time:

a) Time between the ES and schedule start for an activity.
b) Time period acting as a constraint for the beginning of an activity.
c) Time in overlapping related activities between the start of one and the start of the second.


Loop: A networking drafting error situation in which activities are so arranged as to cause a sequence of work to return to the originating point.

Leeway: Synonym for float

Logic diagram: The network plan without quantification time data.

Milestone: Planned date to start or to finish a special event.

Merge: To combine or relate two or more networks at any level to form a single network.

Node: A circle graphically depicting the beginning or the ending event of an activity.

Normal cost: Direct cost estimate for an activity based on the normal time to perform it.

Normal time: The estimated duration time allocated to an activity according to established standards.

Operational planning: That aspect of planning that relates to how things will be accomplished.

Original schedule: The user approved schedule at the time, or immediately following project award.

Output: Material coming out of the computer which results from processing input data

Precedence diagramming: A network diagramming technique for showing the relationships of activities in a project with rectangular boxes symbolizing the activities.

Print out: A document usually produced from a computer that translates data into usable, readable form.

Proposed schedule: A preliminary schedule used for a study and/or proposal.

Remaining Duration: Time required to finished an activity once it has been started.

Revised schedule: Originaladjusted schedule after actual data and /or new change decisions has been introduced.

Resource leveling: The process of scheduling activities within their available float so as to control fluctuations of resource requirements.

Status: Activity completion level at a given point in time.

Sorting: The selective arranging of data to provide a needed pattern of information.

Strategic planning: That aspect of planning which evaluates what must be done within a reasonable period of time.(usually 5 years or more)

Target: A copy of the original plan saved to compare it against performance in the field.

Time compression: A scheduling technique by which project duration is shortened by resource allocation speculation.

Updating: Revising the monitoring plans and documents to reflect current performance, needed changes on logic, durations, and physical resource allocations as of a given date.

Work Breakdown Structure: Breakdown of a project in activities and sub-activities to allow better planning and control.

Working days: Scheduling time units for measuring activity durations which excludes weekends and holidays.


REFERENCES


Manzanera I., PLANNING AND SCHEDULING REFERENCES, Aramco Saudi Arabia, 1991

Lasser's, J.K.,"BUSINESS MANAGEMENT HANDBOOK, McGraw Hill Book Company, 1998

Joseph J. Moder, Cecil R. Phillips, Edward W. Davis, PROJECT MANAGEMENT WITH CPM, PERT AND PRECEDENCE DIAGRAMMING, Van Nostrand Reinhold Company, 1995