Ways to improve performance on projects
Francis M. Webster
The University of Michigan—Flint
An extensive review of the literature of project management reveals a distinct tendency toward “tunnel vision.” One aspect of this is in discussion of ways to improve project performance with respect to schedule. There are two general approaches to this of relevance here, performance to the overall schedule and reducing project duration.
The “tunnel vision” in this respect is the tendency to recognize only a few approaches to the problem of achieving improved project performance, basically those associated with specific network based techniques including PERT, CPPS, and Resource Management. Advocates of PERT stress the “honesty in estimating” argument, one which seems to have lost much of its credibility. CPPS advocates stress the optimum allocation of additional resources, a technique which seems applicable in a relatively limited set of circumstances. Resource Management adovcates point out that schedules which require more resources in a specific time period than are available are not likely to be completed in a timely manner. Nevertheless, this technique has not enjoyed wide spread use for whatever reason.
Experience indicates there are several basic approaches available. The following is a discussion of those approaches indicating their nature, advantages, and disadvantages. They are listed in a specific sequence to suggest the relative desirability of each as to cost and/or management desirability. This sequence is basically a personal judgment so is subject to re-ordering by individual project managers and/or as applied to specific projects.
1. Managerial Attention and Involvement – Every function or activity in an organization argues for more management attention to their needs, yet seldom is this explicitly considered as one of the variable resources of a project. Leadership style varies substantially and has been shown to affect performance in a number of environments. With a few exceptions, when management gives extra attention to an activity, it results in better performance. This can vary from an occasional direct inquiry or visit to the work site to intensive participation in the activity until a major problem has been resolved. For example, a call from the president to stress the importance of a deadline or a superintendent personally driving a fork lift to move a stack of lumber. Both provide stimulus to the person responsible for performing that work. Often, simply demonstrating that a specific task can be done restores determination of the worker to accomplish it. At a lesser degree of participation, knowledge of what is to be done and a few well timed questions as to progress will succeed in stimulating good performance. Sometimes it will be disclosed that the real problem is communication of what is expected in the performance of the activity, i.e., the specification of the work.
Care must be taken here to avoid unnecessary “meddling” which can be detrimental to morale and progress. A few leadership styles are so abrasive as to destroy morale when contacting employees at lower levels in the organization. The sensitive manager will recognize this risk. Unfortunately, often the more abrasive managers are either unaware of the effect of their behavior on others or they really do not care.
The above may imply only top management attention. Actually, the objective is to achieve an appropriate degree of attention at every level in the organization from foreman to supervisor to chief executive officer. Timely direction to proceed is fundamental to avoid lost time at the beginning of a shift or between activities. Sensing inadequacy or misunderstanding of instructions at an early stage is essential. Recognizing conflicts between people or sub-units in the organization can minimize not only time lost but also deterioration in cooperation. Simple coordination of efforts can make work flow smoothly and avoid mistakes which are not only costly in terms of rework but also in terms of worker morale and pride in the work.
The costs of increasing management attention are generally relatively small and the pay-offs are sometimes large. One way to facilitate this action is by having an effective management information system that keeps managers informed of what is happening in their organization.
2. Expediting – Expediting is taking action to assure performance of an activity to schedule in accordance with the existing plan. It often requires only the checking on progress, the removal of an obstacle to progress, or possibly minor changes in method. For example, material delivery can be expedited by checking the location of a freight car, seeing that a bill is paid and thus releasing a credit hold, or changing the mode of transportation to a faster one.
Although seldom discussed explicitly in project management literature, expediting is probably one of the most frequently used approaches to enhancing activity performance. Indeed, much of the time of first line supervision is spent in expediting. Often it is the key to effective utilization of labor and equipment. Failure of essential materials or equipment to arrive on time may result in an entire crew accomplishing nothing for a whole shift. At minimum, some time is lost while alternatives are checked and efforts redirected. Effective dispatching can keep such losses to a minimum but expediting will almost always be necessary.
Expediting may be a part of the active participation of higher levels of management. It may be assigned to the purchasing, scheduling, or other operating function in the organization. Often one or more persons are assigned this as a sole responsibility. For purposes of this analysis, it is not considered to be a direct application of resources to the activity but rather an overhead cost of the organization.
Full advantage should be taken of the management hierarchy in expediting. Only that amount of leverage should be applied that is required to get results. Relying on top management expediting on every occasion will soon make it the only effective approach.
An adequate management information system is essential in the expediting process. Checking on progress must be early enough to allow reaction time prior to delays affecting performance of the activity. Thus, it is often desirable to require reporting of progress prior to the finish of an activity. For example, obtaining shipping data, carrier, and vehicle number on shipments often provides a week of lead time in the event of failure to perform to schedule. Preventing late arrival of scheduled resources is essential to orderly flow of work and avoidance of nonproductive labor and equipment time.
3. Improved Methods – Often, application of Industrial Engineering work design principles can improve the methods used by the workers. This may necessitate making available to the supervision an industrial engineer to help them select the best method. It would also seem to be productive to train supervision in the techniques and principles of work design and methods improvement, including job enrichment, job enlargement and other related management techniques. On-the-job training is not adequate to develop proficiency in these skills.
Often improved methods involve equipment and facilities not presently available in the organization. If the equipment is not available by short term rent or lease it may have to be purchased. This requires a forecast of possible usage over some future period. Economic analysis is required to determine justification. Special training in these procedures may be required. Skills not presently available may be required to effectively use such equipment. On the other hand, an organization without the flexibility and resourcefulness to develop and adapt new methods may be left behind in a competitive environment.
Testing of a new automotive engine threatened delay in its introduction for a year. Careful review of the test requirements indicated that only the cam system was critical. As a result, a standard engine was adapted for testing these moving parts rather than wait for the pattern work to be finished necessary to cast the new blocks.
The engine was introduced on time and proved very successful.
In another instance, a 130 ton precast, purchased concrete beam was called for. Originally it was to be fabricated off site and transported in. A re-evaluation of construction strategy suggested fabrication of the beam at the foot of the columns on which it was to rest. This change of method permitted much more rapid progress on the rest of the building and earlier completion.
4. Reassign Resources — Although many resources, such as carpenters, are treated as homogeneous, any foreman can confirm that their abilities vary substantially for a given task. Often, for a variety of reasons, the resources are not assigned to the set of tasks on which they would be most capable. In this event, the resources can be reassigned to tasks where they are the most qualified and especially on the critical activities. This is basically the classical “assignment problem” of management science. The cost of this can be very low, possibly only some small resentment by those who did not get the activity they wanted or occasionally a delay of an activity until the desired resource is released from working on another activity.
This approach to improving project performance is common in a drafting room. Progress on specific plates is closely monitored by supervision against required completion dates. As other plates are completed, a more experienced person may take over on one that is not going so well in order to meet the deadline. In other instances, the more senior person may start the plate in order to assure accuracy of certain key dimensions, etc.
5. Re-allocate Resources — At certain times during a project there are resources which are not required. Often these can be re-allocated to critical activities with a resulting improvement in performance on the critical activity and therefore on the project. Although doubling the resources on an activity will not normally halve the time required, this is still an economical approach to improving performance. While the cost of performing the activity may be greater, organizational cost will likely be constant. If not reallocated, these resources might otherwise incur costs for which no productive output is realized.
For both re-assignment and re-allocation of resources, it is important that the project manager realize the need for action. Early knowledge of a potential problem and an understanding of the resources required, as well as available, are important contributions of the management information system to permit achievement of good performance on an activity and on the project as a whole.
6. Overlap Activities — Not considering time objectives, the time related cost of money, or deterioration of completed work, perhaps the simplest project plan to manage is the strictly serial set of activities. As more activities are performed in parallel, i.e., concurrently, more coordination and management attention are required. Also, as the activities are overlapped, greater risks are incurred that some work will have to be done over and perhaps even torn out. Note also that for the same resource application and technical performance level, additional time may be required to perform the individual activities involved. Generally, however, these are relatively small costs to pay for improved project performance.
Overlapping was a contributor to the successful engine project mentioned above. As a result of the alternative method for performing the tests, these tests were conducted in parallel with further design and development work rather than in sequence afterwards.
Overlapping activities can be indicated in network based systems in either of two ways. In activity-on-the-arrow (AOA) notation it is necessary to subdivide the activities into two or more arbitrary segments. The segments can then be shown as illustrated in Figure I for two 20 day activities.
Fig. I – Two Activities Overlapped by Segmenting in Activity-on-the-Arrow Notation
In activity-on-the-node notation, this can be done more readily by one of the relationships as shown in Fig. II.
Fig. II — Two Activities Overlapped in Activity-on-the-Node Notation
Thus, it can be seen that AON notation permits overlapping to be indicated graphically and input into a computer system with greater ease.
7. Define the Activities in Greater Detail – No project should be defined in greater detail than necessary to properly manage. Since most activities will require relatively little attention, the majority of the activities in a project can be defined in rather gross terms. As critical activities are identified, it may be desirable to break them into finer detail. As a result, closer scrutiny may lead to improved methods or more overlapping. Again, the cost of this is relatively small, especially if cost accounting can remain at the less detail level.
On a major management systems project, the original network schedule showed an earliest completion about one and one half years after the commitment date. Review of activities revealed that many activities on the critical path had significant clerical and administrative work content. These were defined in greater detail, identifying specifically those clerical and administrative tasks as opposed to more technical tasks that were essential to project progress. As a result, the critical path was reduced to permit an acceptable completion date for the project.
8. Delete Certain Activities — In every project there are certain activities that are included primarily to provide control, to provide quality and reliability assurance, or that are considered desirable to be done concurrently with other work, i.e., program documentation concurrent with programming and debugging. In practice, we know that these are often delayed until after the primary objective is achieved, or deleted entirely. For quality and reliability assurance type activities, there may be some additional risk associated with deleting the activity or perhaps an additional increment of unit cost if it results in overdesigning the product. For the activity delayed, there may be some additional cost associated with performing it at a later date. Generally, these are not large costs except perhaps when a product is unknowingly underdesigned resulting in excessive warranty or maintenance costs.
9. Change the Technology Applied — Ideally, the technology originally selected to perform an activity or inherent in the product design will be one that is most economical given the tolerance and performance requirements, volume, availability of that technology in or to the organization, etc. If it is necessary to perform the activity in less time, with greater quality, or to improve product performance, another technology may be used at somewhat greater cost. Contrary to some of the prior approaches discussed, the costs associated with changing technology are more subject to quantitative evaluation. Often they will involve obtaining new equipment and, as in many of the DOD/NASA projects, the actual invention of that new technology.
A change in technology applied contributed to the completion of the Pontiac Silverdome on time and within budget. When a strike threatened delays in delivery of structural steel, the design was changed to poured-in-place concrete columns and beams. The result was not only efficient from design and construction points of view but was also aesthetically pleasing.
10. Sub-Contract or Buy — If the resources are not available within the organization, it may be necessary to call upon outside sources to accomplish an activity. This approach may permit improvements in performance in time, resource usage, and/or technical performance depending on the nature of the work, the organization’s capabilities, and the supplier’s capabilities. This may be an alternative when, for example, the original plan relied on internal resources to keep them employed. Sometimes relying on outside suppliers may result in cost increases, more coordinative efforts, and possibly technical problems but may also result in improvements in these dimensions.
Normally, most of the maintenance work on machine tools is performed in-house. The year that collapsible steering columns were introduced, a large number of machines had to be repaired, overhauled, or rebuilt. The volume of work required that much of it be sent out to outside contractors. Only in this manner could the strict government deadlines be met.
11. Applying Additional Resources — Perhaps the most obvious means of improving performance on activities, and the one to which appeal is most frequent, is adding resources. This may vary from some overtime, to extra personnel and/or equipment, to going to multi-shift operation.
Overtime is often the first choice of the workers due to increased compensation. It is also the first choice of management due to absence of hiring, firing, training, fringe benefits, and other costs as well as its ease of implementation. There are physical limits on this alternative, however, beyond which it is necessary to add units of resource. Productivity decreases as the amount of overtime approaches these physical limits.
The productivity of additional resources will also tend to be less, assuming the original allocation was reasonably near optimum. To the extent that additional resources can be profitably utilized over an extended period of time, this reduction in efficiency may be the only significant cost. To the extent they are required for only a relatively short period of time, there may also be appreciable costs of hiring, training, and firing. Thus, the cost to the organization could be substantial.
Caution must also be exercised in using this approach due to its possible effect on the local labor market. A major stamping plant was being built at a time when there was considerable other construction activity in the area. Attempts to accelerate construction of the stamping plant resulted in increased effective wage rates due to necessity of work overtime, inability to deal with a slower workpace by individual workmen, and decreasing competence of new hires. As a result excessive overtime was incurred on that project.
This approach is the basic concept of the time-cost trade-off analysis of Critical Path Planning and Scheduling (CPPS) as developed by Kelly and Walker. It is also frequently the first suggested by the performers of the activity. Compared to the previously mentioned alternatives, it may be very expensive and more so to the performing organization than is suggested by the CPPS analysis.
12. Change Target Dates — This and the next approach may be the least desirable because they require the client’s agreement, or at least the ignoring of his desires. A decision to incur a small delay may result in a whole year delay in announcing a new product, a delay which could be quite serious. There may be a penalty clause associated with late delivery and the client may be unwilling to use the firm’s services again. On the other hand, occasionally the client may have reasons for being willing to accept a delay to avoid cash outflows of his own. Although changing the target date may seem to be allowing activity performance to degenerate, in terms of the other dimensions of an activity, i.e., resource application rate and technical performance, the net result may be an improvement. This could be true if incessant pressure to meet target dates leads to short cuts in performing the activity resulting in lower technical performance or even extreme frustration of the participants and deterioration of motivation to complete the activity at all, let alone meet the performance criteria.
In one case, an estimated one million dollars was saved by delaying a target date by one month. In the construction of a major casting plant, the architect-engineering firm proposed a one month delay in publication of the bid specification package. They estimated that an additional five million dollars of work could be incorporated in that month. A check of the network indicated no major scheduling problems would ensue. Thus, the delay was granted. It was estimated conservatively that a 20% premium would be paid for that work if it were let outside of the general bid package.
13. Change the Scope of the Project — This may be a more acceptable alternative to the client than changing the target date. Sometimes the client may be able to use a portion of a building. A product announcement may be made with a prototype or a decision made without complete answers. Candid and persuasive discussion with the client may well result in acceptable alternatives to all parties. It is at this point that the negotiating skill of the project manager can be most valuable. Being able to present clear facts to the client will aid in gaining the client’s acceptance.
14. Change the Person Responsible — While this is frequently used, it is not very desirable from the point of view of the person concerned or the organization. However, sometimes it is an essential action.
When it happens, there will inevitably be delays and other disturbances in the project. The new person will have to learn the past, present, and future of the project. Until this is done, decisions will be delayed or of questionable quality. In the latter case, rework may result and in either case, a succession of perturbations can be expected for some time into the future. Furthermore, if the need for change is not widely recognized and accepted in the project organization, it can create anxieties and other behavioral disturbances, some as severe as people resigning from the project team.
15. Change the Management Information and Control System — This is generally done with great reluctance due to the lead time for implementation. If a new computer is required there is lead time for installation. If a new system must be designed and installed, months if not years may pass before it is ready for use. Even if existing software is purchased, there is a lead time for getting it to work effectively on a different computer. All of these imply costs. In addition, an often ignored cost is that of training existing personnel in new methods and gaining their supportive use of the new system.
There is no substitute for clearly assigned responsibility. It is vital, however, that a management information and control system be provided that will assist those who have been assigned responsibility. It is essential that they be provided the right information at the right time and in the right format that will allow them to effectively manage the project.
16. Change the Organizational Structure — This is a very severe action and is often a last resort. Its ramifications are wide spread. It often is accompanied by major personnel changes. New working relationships must be established. The management information systems may have to be modified. Old commitments within the organization may have to be renegotiated if not voided altogether. Since these things seldom happen instantaneously at all levels in the organization, it may take months for a return to stable equilibrium and efficient and effective performance.
The Director of a Community Development Department of a city determined that the department would be more effective if it were re-organized from a traditional functional arrangement to a matrix concept. He wondered why the change was not working. Analysis showed three serious defects in his plan. First, although a consultant was developing a network based system to support the project managers, it was not in place yet. The organization was still dependent upon the management information system designed to support the functional organization. Second, the job descriptions had not been revised to reflect the changes in responsibilities of the members of the organization. Third, the “project managers” had very limited responsibilities, much more consistent with those of a project coordinator or expediter. Whether these deficiences were corrected or not is not known but the option of changing the responsible person may have been exercised.
The literature of project management is very limited in discussion of ways to improve performance. Sixteen specific ways are enumerated in an order judged by the author to represent their increasing cost and/or decreasing managerial desirability. The most significant observation is that applying additional resources is actually eleventh in this list
Undoubtedly, more could be said about the advantages and disadvantages of each as well as appropriate caveats in their application. The discussion presented is intended to elicit further thought by the reader. The author would be pleased to exchange ideas with anyone interested in expanding on this list or discussion of specific items.
PROJECT and BUSINESS SYSTEMS
Pragmatic cost saving
Sparks, NV 89431
Message relay —