Improvement Portfolio and Resource Capacity

Why it matters

A project can be useful and feasible on its own; when it is selected together with other projects that want the same expert, the plan does not work. An improvement portfolio addresses this question of running them together. Which projects are chosen matters as much as in what order they will be done and which will wait. Starting many jobs without a link to the strategic goal accumulates work in progress instead of completed results. The resource limit is not only the budget. The expert who can take measurements, maintenance time, verification space and the time employees can set aside for learning can also be limits. In this lesson the work is done on shared expert hours; the suitability of other resources is taken as checked separately.

A single project score can organize the priority discussion; it is not proof of feasibility. A job whose impact looks large by score may first require the data definition to be corrected or a safe implementation to be verified. Even if a precondition has a low score of its own, the later result cannot be obtained without it. The resource load is the sum not only of the eye-catching projects but also of the necessary preconditions. If the same precondition serves two projects it is counted once; if it has to be redone for each project, the jobs are defined separately. When this distinction is not clear, the cost can be calculated both higher and lower than needed. Dependencies must be made visible before the resource table.

The same care is needed on the impact side. If two projects rescue the same late orders, their contributions cannot be added directly. One project may be what makes the other's effect possible; in that case writing a separate result gain for the precondition can create a double count. In another case the scopes of the projects are really separate and the contributions can be added. The decision rests on the shared result definition and the relationship at the record level. Do not read a score total as a delivery gain or a money saving. A score is a ranking aid; its unit is not orders or liras. The expected result is reported separately from the realization to be measured after implementation. This distinction is necessary so that the portfolio can be evaluated honestly afterward.

Portfolio selection is not limited to finding the highest total. On-time completion, uncertainty and the conditions to protect also affect the decision. Committing all the hours of a period in advance can stop critical work if a small estimate deviation occurs. If time needs to be left in reserve, this choice is explicitly subtracted from capacity; it is not kept as a hidden resource outside the calculation. The impact, load and risks of two feasible sets are compared in the same table. A set with lower impact can be more reliable under some conditions. But declaring the low-impact option better without writing the rationale is not right either. The decision must be bounded by the given constraints and the evidence at hand.

Formula