GM592 Unit 6 Planning Fallacy Seminar Reflection Example

Six Weeks Planned, Nineteen Taken: A Unit 6 Seminar Reflection on the Line Nobody Challenged

[Student Name]

Purdue University Global

GM592: Project Planning and the Project Plan

Unit 6 Seminar Reflection

[Instructor Name]

[Date]

The sawmill, the classmates and the mills described are composite; the research cited is real.

The Log Deck

Two years ago at Tygart Ridge Hardwoods, we replaced the log deck that feeds the headrig. I planned six weeks. It took nineteen. At the time I blamed three causes: the steel fabricator delivered the deck frame five weeks late, the concrete footings had to be redone after an inspection, and a wet spring kept the crane off the yard for two weeks. I told the owner it was bad luck, mostly the fabricator.

The Other Cases

Classmates brought similar stories. Kyle, a software lead, described a feature his team estimated at two weeks that took eleven, with each delay explained by a different surprise. Nora, an electrical estimator, said her company's bids ran long every winter, and that every winter the estimators blamed the weather as if it were new. Listening to them, I noticed that each of us had a separate excuse for each delay and no explanation for why we were always late.

The Question

Then Hector, who manages a bottling plant, asked me about my current project. I had described the kiln and mentioned the eight-day cycle for 4/4 red oak. He asked, "How long until the new kiln actually runs at eight days?" I said it would run at eight days from the first saleable charge. He said that every new line in his plant had taken months to reach its rated speed.

I checked my estimate worksheet that evening. It used ten days for the trial charge, but every charge after that was planned at the vendor's nominal cycle. No one had questioned that line, including me.

My worksheet assumed the kiln would dry at its rated cycle from the first saleable charge, an assumption that every other new installation described in the seminar contradicted.

What the Research Explains

Daniel Kahneman and Amos Tversky argued that people forecasting a project focus on its specific plan and neglect the distribution of outcomes for similar projects, and that forecasts improve when they start from that wider class (Kahneman & Tversky, 1979). Bent Flyvbjerg turned the idea into a method he calls reference class forecasting: gather the real outcomes of projects like yours, then move your own estimate toward them (Flyvbjerg, 2006).

That described my log deck exactly. I had planned from the inside, step by step, and treated each delay as an accident. Roger Buehler, Dale Griffin and Michael Ross found that people predicting their own task times focus on plans for the future and discount their past experience, even when they remember being late before (Buehler et al., 1994). I had been late on every yard project I could name. I had never asked how long log decks at other mills had taken.

So I looked for a reference class for the kiln. The state hardwood association's newsletter has carried accounts from three mills that installed package kilns in the last five years. All three reported that their first season's cycles ran 20 to 40 percent longer than the vendor's nominal schedule, while operators learned the controls and adjusted schedules for their own lumber. Hector's question had a factual answer, and it was not the one my worksheet assumed.

Why the Fabricator Took the Blame

Looking back at the log deck, the fabricator's five weeks were real, but they were not the reason I was wrong. If the fabricator had been on time, the project would still have taken about fourteen weeks, more than twice my estimate. The footings, the weather and a dozen smaller problems would still have happened. Blaming the fabricator let me keep my method, because it made the overrun look like one supplier's failure rather than the normal result of planning without a reference class. The seminar made me see that a list of excuses, one for each delay, is itself a sign of inside-view planning, since the outside view would have expected some delays without knowing in advance which ones.

What Changed

I have revised the worksheet. The first three saleable charges now carry a cycle of about 11 days, 40 percent above nominal, and the following five carry about 10 days. Only after that does the plan assume eight. The change moves the date when the kiln can supply the cabinet maker's full monthly volume by about three weeks, and I have told the sales manager before the customer hears a date. I also added a new column to the worksheet: for every duration, the reference class it was checked against, or the word none.

References

Buehler, R., Griffin, D., & Ross, M. (1994). Exploring the "planning fallacy": Why people underestimate their task completion times. Journal of Personality and Social Psychology, 67(3), 366-381. https://doi.org/10.1037/0022-3514.67.3.366

Flyvbjerg, B. (2006). From Nobel Prize to project management: Getting risks right. Project Management Journal, 37(3), 5-15. https://doi.org/10.1177/875697280603700302

Kahneman, D., & Tversky, A. (1979). Intuitive prediction: Biases and corrective procedures. TIMS Studies in Management Science, 12, 313-327.