PM 570 Week 2 The Project Life Cycle Example

Gates, Prototypes and a Certification Lab: Designing the Life Cycle for a Refrigerant Conversion Project

[Student Name]

University of Phoenix

PM/570: Projects, Programs, and Portfolios

Week 2 Assignment

[Instructor Name]

[Date]

Red River Air Systems, its project phases, dates and figures are composites written for a model paper.

Week 1 introduced Red River Air Systems, an invented Wichita Falls, Texas, maker of light commercial rooftop units, and the cross-functional team converting its three-to-five-ton line from R-410A to R-454B, a mildly flammable refrigerant with much lower global warming potential. The project has a budget of $6.8 million and about 15 months to reach production before the regulatory deadline after which the old design can no longer be sold for new installations. This paper designs the project's life cycle: the stages, the gates between them, the iteration inside development and the controlled overlap that makes the schedule possible.

Why a Gated Life Cycle With Iteration Inside

Two features of the project shape its life cycle. The date is fixed by regulation, and the product must meet a safety standard that requires refrigerant detection and mitigation for mildly flammable refrigerants. Both argue for clear decision points where evidence is reviewed before money is committed. At the same time, the design contains real unknowns: how much refrigerant charge the coils need, where leak sensors detect reliably and how controls should respond. Those unknowns are best resolved by building and testing, not by analysis alone.

Cooper (2008) described how stage-gate systems had evolved to become more flexible, allowing stages to overlap, adapting gates to project risk and incorporating iterative build, test and feedback loops within stages. Thomke (1998) showed through studies of product development that the way firms structure experimentation, including the speed and cost of each test cycle, affects development cost and time. Red River's life cycle combines both ideas: firm gates for commitment decisions and fast prototype loops for learning.

Stage 1: Scoping

Deliverables: refrigerant selection confirmed, regulatory and certification requirements listed, competitor analysis and a first estimate of cost and schedule. Gate 1 criteria: refrigerant availability from two suppliers, a feasible path to certification and executive agreement that the project fits the conversion strategy.

Stage 2: Business Case and Plan

Deliverables: product requirements, including capacity, efficiency targets and cost ceiling; a project plan; a supplier strategy for compressors, sensors and valves; and a plant change assessment. Gate 2 criteria: requirements signed by engineering, manufacturing, quality and sales; supplier letters of intent; and a schedule that reaches production at least eight weeks before the compliance date.

Stage 3: Development With Prototype Loops

Development runs three planned loops, each about four weeks. Loop A tests refrigerant charge and coil performance on two prototypes in the company's test chambers. Loop B tests leak sensor placement and response, deliberately releasing small amounts of refrigerant in a controlled enclosure to confirm detection times. Loop C integrates controls, testing the mitigation sequence that starts the blower and shuts the compressor when a leak is detected. Each loop ends with a short review in which results change the design. A fourth loop is held in reserve if results require it.

Sommer and Loch (2004) analyzed projects facing complexity and unforeseeable uncertainty and argued that trial-and-error learning and parallel trials can be more effective than planning alone when the problem cannot be fully specified in advance. The prototype loops are a deliberate use of learning where planning cannot settle the answers.

The gates decide whether to spend; the loops decide what to build.

Stage 4: Testing and Validation

Deliverables: certification testing at an independent laboratory, reliability testing, a pilot production run on the modified line and dealer field trials at six sites. Gate 4 criteria: certification test reports with no open nonconformities, pilot first-pass yield of at least 95 percent, field trial units running for 60 days without safety events and service procedures validated by the trainer.

Stage 5: Launch

Deliverables: full production, dealer training completed in all regions, service parts stocked and marketing materials released. The project closes when production has run at target yield for four weeks and the first 1,000 units have shipped.

Overlapping Certification and Tooling

The baseline sequence would finish certification testing before ordering production tooling for the new coil and cabinet changes. That sequence ends about five weeks after the internal target. Overlapping the two saves about seven weeks but risks rework if certification finds a problem requiring a design change after tooling is cut.

Krishnan et al. (1997) modeled overlapping product development activities and showed that whether overlap is worthwhile depends on how quickly the upstream information settles and how sensitive the downstream activity is to changes in it. Applied here, the question is whether the design features that tooling depends on, coil dimensions and cabinet openings, are stable before certification ends. Loop A and Loop C fix those features; certification is more likely to require changes to controls software or sensor mounting, which do not affect tooling. The project therefore allows overlap under three conditions: coil and cabinet dimensions frozen after Loop C, preliminary laboratory tests showing no structural issues and a tooling contract that prices modifications in advance.

Connecting the Life Cycle to Strategy and Operations

The life cycle ends in operations rather than at a handover. Dealer training and service readiness are launch deliverables because the strategic goal is to keep and win dealers, and dealers judge a new refrigerant largely by whether their technicians are ready to handle it safely. Plant readiness is built into Stage 4 through the pilot run so that the line is proven before volume production. The project's closure criterion, four weeks of stable production, ensures that the operations team receives a working process rather than an unproven one. Beyond this project, the life cycle becomes a template for the other product lines that must convert, which leads to the program structure examined in Week 3.

What the Life Cycle Costs

The structure is not free. Each prototype loop costs about $140,000 in materials, chamber time and engineering hours, and each gate review takes roughly two days of senior managers' time to prepare and hold. A simpler life cycle with fewer loops would save perhaps $280,000 but would push learning about charge, sensors and controls into certification testing, where every failure costs a laboratory slot and weeks of delay. The project manager presented both options at Gate 2, and the review board chose the loops as the cheaper way to buy certainty.

Governance Through the Life Cycle

Each gate is decided by a review board of the vice presidents of engineering, operations and sales, with the quality director holding a veto on safety criteria. Gates are scheduled in advance, and the project manager presents evidence against the written criteria. A gate can approve, approve with conditions, send the project back for more work or stop it. The overlap decision is a formal condition at Gate 3, recorded with its rationale.

Conclusion

Red River's conversion project uses a gated life cycle for the decisions that commit money and protect safety and fast prototype loops for the learning that analysis cannot provide. Controlled overlap of certification and tooling, justified by research on when overlap pays, recovers the time the regulatory date demands. The life cycle ends in stable production and ready dealers, the outcomes that protect the company's largest product line.

References

Cooper, R. G. (2008). Perspective: The Stage-Gate idea-to-launch process: Update, what's new, and NexGen systems. Journal of Product Innovation Management, 25(3), 213-232. https://doi.org/10.1111/j.1540-5885.2008.00296.x

Krishnan, V., Eppinger, S. D., & Whitney, D. E. (1997). A model-based framework to overlap product development activities. Management Science, 43(4), 437-451. https://doi.org/10.1287/mnsc.43.4.437

Sommer, S. C., & Loch, C. H. (2004). Selectionism and learning in projects with complexity and unforeseeable uncertainty. Management Science, 50(10), 1334-1347. https://doi.org/10.1287/mnsc.1040.0274

Thomke, S. H. (1998). Managing experimentation in the design of new products. Management Science, 44(6), 743-762. https://doi.org/10.1287/mnsc.44.6.743