430 CHAPTER 9 / RESOURCE ALLOCATION

BIBLIOGRAPHY

Adler, P. S., A. Mandelbaum, V. Nguyen, and E. Schwerer. Getting the Most Out of Your Product Development Process. Harvard Business Review, MarchApril 1996.Corwin, B. D. Multiple R and D Project Scheduling with Limited Resources. Technical Memorandum No. 122,Department of Operations Research, Cleveland: Case Western Reserve University, 1968.Davis, E. W., and J. H. Patterson. A Comparison of Heu-ristic and Optimum Solutions in Resource-Constrained Project Scheduling. Management Science, April 1975.Gilyutin, I. Using Project Management in a Nonlinear Envi-ronment. Project Management Journal, December 1993.Goldratt, E. M., and J. Cox. The Goal. Great Barrington, MA: North River, 1992.Goldratt, E. M. Critical Chain. Great Barrington, MA: North River, 1997.Harrison, F. L. Advanced Project Management. Hants, Great Britain: Gower, 1983.Ibbs, C. W., S. A. Lee, and M. I. Li. Fast-Trackings Impact on Project Change. Project Management Journal,December 1998.Kurtulus, I., and E. W. Davis. Multi-Project Scheduling: Categorization of Heuristic Rules Performance. Manage-ment Science, February 1982.

Kurtulus, I., and S. C. Narula. Multi-Project Scheduling: Analysis of Project Performance. IEEE Transactions on Engineering Management, March 1985.Mantel, S. J., Jr., J. R. Meredith, S. M. Shafer, and M. M. Sutton. Project Management in Practice, 4th ed. Hoboken, NJ: Wiley, 2011.McCarthy, J. Dynamics of Software Development. Red-mond, WA: Microsoft Press, 1995.Nicholas, J. M. Managing Business & Engineering Proj-ects. Englewood Cliffs, NJ: Prentice Hall, 1990.Piney, C. Critical Path or Critical Chain: Combining the Best of Both. PM Network, December 2000.Raz, T., R. Barnes, and D. Dvir. A Critical Look at Criti-cal Chain Project Management. Project Management Journal, December 2003.Trietsch, D. Why a Critical Path by Any Other Name Would Smell as Sweet. Project Management Journal,March 2005.Weist, J. D. A Heuristic Model for Scheduling Large Projects with Limited Resources. Management Science,February 1967.Wheatley, M. Future Perfect. PM Network, April 2004.

C A S ED. U. SINGER HOSPITAL PRODUCTS CORP.

Herbert F. Spirer

D. U. Singer Hospital Products Corp. has done suf cient new product development at the research and develop-ment level to estimate a high likelihood of technical success for a product of assured commercial success: A long-term antiseptic. Management has ins tructed Singers Antiseptic Division to make a market entry at

Copyright Herbert F. Spirer. Reprinted by permission.

the earliest possible time; they have requested a com-plete plan up to the startup of production. Marketing and other plans following startup of production are to be prepared separately after this plan has been completed.

Project responsibility is assigned to the divisions Research and Development Group; Mike Richards, the project scientist who developed the product, is assigned responsibility for project management. Assis tance will

The following case describes the evolution of a new product and the project devised to take it to market. As well as discuss-ing the issues of developing a work breakdown structure, network diagram, schedule, and resource loading diagrams for each of the involved departments, the case also brings up the issues of timecost trade-offs, cash ows, and resource leveling.

CASE 431

be required from other parts of the company: Packag-ing Task Force, R & D Group; Corporate Engineering; Corporate Purchasing; Hospital Products Manufacturing Group; Packaged Products Manufacturing Group.

Mike was concerned about the scope of the project. He knew from his own experience that a nal formula had yet to be developed, although such development was really a routine function. The remaining ques-tions had to do with color, odor, and consistency addi-tives rather than any performance-related modi cation. Fortunately, the major regulatory issues had been resolved and he believed that submission of regulatory documentation would be followed by rapid approval as they already had a letter of approval contingent on nal documentation.

But there were also issues in packaging that had to be resolved; development of the packaging design was one of his primary concerns at this time. Ultimately, there will have to be manufacturing procedures in accordance with corporate policies and standards: capital equipment selection and procurement, installation of this equip-ment and startup.

Mike was concerned about de ning the project un-ambiguously. To that end, he obtained an interview with S. L. Mander, the group vice-president.

When he asked Mander where his responsibility should end, the executive turned the question back to him. Mike had been prepared for this and said that he would like to regard his part of the project as done when the production process could be turned over to manu-facturing. They agreed that according to Singer practice, this would be when the manufacturing operation could produce a 95 percent yield of product (fully packaged) at a level of 80 percent of the full production goal of 10 million liters per year.

But I want you to remember, said Mander, that you must meet all current FDA, EPA, and OSHA regu-lations and you must be in compliance with our internal speci cationthe one Ive got is dated September and is RD78/965. And you know that manufacturing nowquite rightly, I feelinsists on full written manufactur-ing procedures.

After this discussion, Mike felt that he had enough in-formation about this aspect to start to pin down what had to be done to achieve these results. His rst step in this ef-fort was to meet with P. H. Docent, the director of research.

You are naive if you think that you can just start right in nalizing the formula, said Docent. You must rst develop a product rationale (a).* This is a for mally de ned process according to company policy. Market-ing expects inputs at this stage, manufacturing expects their voice to be heard, and you will have to have ap-provals from every unit of the company that is involved; all of this is reviewed by the Executive Committee. You should have no trouble if you do your homework, but expect to spend a good eight weeks to get this done.

That certainly stretches things out, said Mike. I expected to take 12 weeks to develop the ingredient formula (b) and you know that I cant start to establish product speci cations (c) until the formula is complete. Thats another three weeks.

Yes, but while you are working on the product spec-i cations you can get going on the regulatory documen-tation (d). Full internal speci cations are not required for that work, but you cant start those documents until the formula is complete.

Yes, and I nd it hard to believe that we can push through both preparation of documents and getting approval in three weeks, but Environmental swears it can be done.

Oh, it can be done in this case because of the prepa-ratory work. Of course, I wont say that this estimate of three weeks is as certain as our other time estimates. All we need is a change of staff at the Agency and we are in trouble. But once you have both the speci cations and the approval, you can immediately start on developing the production processing system (g).

Yes, and how I wish we could get a lead on that, but the designers say that there is too much uncertainty and they wont move until they have both speci cations and regulatory documentation and approval. They are offer-ing pretty fast response; six weeks from start to nish for the processing system.

They are a good crew, Mike. And of course, you know that you dont have to delay on starting the pack-aging segment of this project. You can start developing the packaging concept (e) just as soon as the product rationale has been developed. If my experience is any judge, it will take a full eight weeks; youll have to work to keep the process from running forever.

But as soon as that is nished we can start on the design of the package and its materials (f), which usually takes about six weeks. Once that is done we can start de-veloping the packaging system (h), which shouldnt take longer than eight weeks, concluded Mike. At this point he realized that although Docent would have general

*Tasks which must be accounted for in a network plan are iden-ti ed by lower-case alphabetic symbols in parentheses. Refer to Exhibit 1.

432 CHAPTER 9 / RESOURCE ALLOCATION

knowledge, he needed to talk directly to the Director of Manufacturing.

The rst step, which follows the completion of the development of processing and packaging systems, said the Director of Manufacturing, is to do a com-plete study of the facilities and equipment requirements (i). You wont be able to get that done in less than four weeks. And that must precede the preparation of the capital equipment list ( j) which should take about three-quarters as long. Of course, as soon as the develop-ment of both the process system and packaging system are completed, you could start on preparing the written manufacturing facilities procedures (q).

But, said Mike, Can I really nish the procedures before I have installed the manufacturing facilities (p)?

No, quite right. What you can do is get the rst phase done, but the last three of the ten weeks it will take to do that will have to wait for the installation of the manufac-turing facilities.

Then this means that I really have two phases for the writing, that which can be completed without the manufacturing facilities installation (q), and that which has to wait for them (q).

True. Now you realize that the last thing you have to do after completing the procedures and installing the equipment and facilities is to run a pilot test (r) which will show that you have reached a satisfactory level?

Yes. Since that must include debugging, Ive estimat ed a six-week period as adequate. The director of manufac-turing assented. Mike continued, What Im not sure of is whether we can run all the installation tasks in parallel.

You can let the purchase orders and carry out the procurement of process equipment (k), packaging equipment (l), and facilities (m) as soon as the capital equipment list is complete. The installation of each of these types of equipment and facilities can start as soon as the goods are on hand (n, o, p).

What do you estimate for the times to do these tasks? asked Mike. The director of manufacturing es-timated 18, 8, and 4 weeks for the purchasing phases for each of the subsystems in that order and four weeks for each of the installations. Then I can regard my job as done with the delivery of the procedures and when I show my 95 percent yield, said Mike, and the director of manufacturing agreed, but reminded Mike that none of the purchasing cycles could start until the capital equipment list had been prepared and approved ( j) which he saw as a three-week task.

The executive committee of D. U. Singer Hospi-tal Products Corporation set a starting date for the

project of March 10 and asked Mike to project a completion date with his submission of the plan. The committees request implied that whatever date Mike came up with was acceptable, but Mike knew that he would be expected to show how to shorten the time to complete the project. However, his task in making the schedule was clear; he had to establish the resource requirements and deal with calendar constraints as best as he could.

To this end, Mike had to get an estimate of resources, which he decided to do by making a list of the acti vities and asking each group involved what was their level of employee input. The results of this survey are shown in Exhibit 1. For example, activity a takes 8 weeks and requires 12 worker-weeks from R&D, or an aver-age of 1.5 workers for the entire 8-week duration of activity a.

For the purposes of overall planning, the accounting department told Mike that he could estimate a cost of $600 per week per employee. This would enable him to provide a cash ow forecast along with his plan, which the chief accountant said would be expected, something that Mike had not realized.

Mike knew that it was customary at D. U. Singer to provide the following as parts of a plan to be submitted to the executive committee:

A. Statement of objectives.B. Work breakdown structure.C. An activity-on-node network.D. A determination of the critical path(s) and the dura-

tion along the path.E. An activity list, early-start schedule, slack list, and

master schedule. Assume that every activity begins at its early start, regardless of resource constraints.

F. A period labor requirements table for each group and the project as a whole.

G. A cumulative labor requirements table for each group and the project as a whole. Include line graphs to illustrate the cumulative loads.

H. A schedule based on the best leveling of labor requirements that could be achieved without length-ening project duration by more than 14 percent in calendar days.

I. A cash ow requirements graph for the project, assuming that charges are uniformly distributed throughout the activity.

QUESTIONS 433

QUESTIONS

1. Construct the nine elements of the plan identi ed above. 2. Analyze the plan for potential problems. 3. Analyze the plan for opportunities. 4. Should the executive committee approve the plan? Why

or why not?

5. What alternatives might the executive committee sug-gest for analysis?

aprod. rationale 1 12 1 1 2 0 0 $ 0bdev. formula 0 16 4 2 0 0 0 500cprod. spec. 1 6 3 1 1 0 1 0dreg. document 0 12 4 2 0 0 0 0edev. pkg. concept 12 8 4 2 8 0 2 4000fdesign pkg. 12 2 3 0 3 0 3 2000gdev. proces. sys. 0 18 12 12 0 0 0 0hdev. pkg. sys. 24 8 8 0 8 0 2 0istudy facil./eqpt. req. 0 4 16 2 2 0 0 0jcapital equip. list 0 1 3 0 0 0 1 0kprocure proces. eqpt. 0 1 1 1 0 0 7 40,000lprocure pkg. eqpt. 1 0 1 0 1 0 9 160,000mprocure facil. 0 0 1 1 1 1 6 30,000ninstall proces. eqpt. 0 2 4 8 0 4 1 4000oinstall pkg. eqpt. 2 0 4 0 8 4 1 8000pinstall mfg. facil. 0 0 5 5 5 10 1 6000q,qwritten procedures 5 5 5 10 15 10 0 5000rpilot test 3 6 6 6 6 6 0 0

EXHIBIT 1 Labor Requirements (Worker-weeks)