8.1 Converting Customer Needs to Quantitative Metrics

a

8.2 Metrics Effecting Multiple Subsystems

There are some Customer Needs that probably effect multiple subsystems – cost, weight, size, etc. You should NOT map one metric to multiple subsystems because this approach very quickly causes all your subsystems to exceed the Niemi Number. Niemi number exceeding ~8 has the effect of invalidating your down-selection matrix as it’s no longer sensitive enough to differentiate concepts. Each Metric maps to one sub-system.

So, how to deal with situations where one metric very clearly influences two different subsystems or even the whole design? There are two approaches to consider. First, go back to the original Customer Need that spawned the metric that influences multiple subsystems. Ask yourself: are there really two different Metrics here, one influencing Subsystem A and the other influencing Subsystem B? If yes, generate the second Metric and assign the Metrics to their individual subsystems. In this way, the influence of the Customer Need is felt appropriately across different subsystems.

If there are no additional Metrics the Customer Need could spawn to spread its influence, the second approach is to assign the Metric to the subsystem where it has the most influence over the whole design. Consider the weight of your grinder. My sense from taking apart paper shredders is the grinder head (usually solid metal) is going to be the heaviest component. So, assign the weight Metric to the grinder subsystem where the influence of reducing weight has the greatest impact on the weight of the overall product. If the grinder head is 70% of the weight, the housing is 25%, and the electronics are 5%; a grinder head weight reduction of 10% results in 7% weight reduction of the whole system. By contrast, a 10% weight reduction of the other subsystems reduces overall system weight by only 2.5% and 0.5% respectively.