Risk measurement, system boundaries

The phase boundaries of the Pd-X (X = H,D,T) system were determined from pressure concentration temperature data because of the high risk of handling PdT samples outside our tritium loading equipment. Pd forms no stable oxide layers as is the case for or Nb that prevent the tritium to leave the sample. The boundaries between the miscibility gap and the 3-phase were obtained from the shape of the desorption isotherms. The values of concentration and temperature of the solvus line between the a- and the two phase regions a+3 were obtained by quasi isochoric measurements. A PdX sample with the concentration x slightly in the miscibility gap was heated in small temperature steps so that the concentration of the sample decreased and finally belonged to the pure a-phase. The change of slope in the equilibrium pressure as a function of the inverse temperature is interpreted as the intersection with the solvus line. 

The choice of what is appropriate will depend on many factors and the search is to find which level, or which mix, provides the best way of defining and measuring risk control. What are the boundary markers we can most unambiguously define to indicate that the system is moving outside the envelope of control and needs corrective or punitive action ... 

The presented example shows that it is beneficial for a medical device manufacturer to think in a risk-based way about interoperability and integration during device development or alternatively in a separate device integration interface project. Measures that can be applied during development and within extended boundaries and responsibilities of the device and its external interfaces may reduce risks and thus costs for an integrator. Although some measures (like HMAC-based authentication) will have to be implemented by the connected systems too, the usage of proven and standard solutions and libraries reduces the effort considerably. 

---