Building variant products in software defined systems can be used for market position both by offering different features, at different price points, but also by offering different service levels. Typically service levels define and offer minimum or expected goals for ‘availability’, ‘reliability’, or ‘performance’ say, and offer solutions to issues like failures, warranty, or problem management.
For these service level agreements (SLAs), prior measurements are required to understand the potential business benefit, and ongoing measurements to ensure the appropriate quality of service, or even use/abuse by the user. Software-based control systems are well placed to collect and inform the business and provide the ongoing monitoring.
Information from monitoring systems is routinely used for both Condition-based maintenance and Predictive maintenance in many industries in large electrical machines, jet engines and automotive components. These enable optimal service periods, replacement of marginal components to minimise future shop visits, improving availability and even initiate service events to mutual benefit of customer and service shop. The monitoring information can be used as feedback to inform the designers of usage scenarios, the business of excursions outside the warranty or SLA and the service bay with diagnostic information, both live and historic, in support of fault investigation.
The business trick is to use the existing control system sensors to maximise information or its synthesis, and only augment this sensing set when there is clear business value to be gained. Additional instrumentation, its signal processing, the transmission of information and its analysis all add complexity, cost (both unit and recurring) and delay.
The more connected our social world, the greater the opportunity for making this information more immediately available, carry more contextual information (e.g. time, location, user demands, even weather!) and enable businesses to refine their offerings, more value opportunities, or even additional products.
Technically, the key problems are collecting valid (and validated), accurate, information (differentiating signals from noise) without losing valuable transients, local processing, compression or selective storage, opportunities for (robust and secure) transmission which balance immediacy, bandwidth, and cost and off-line analysis of what potentially could be large, continuous, streams of individually correlated, but systematically uncorrelated, data.
For exceptional systems, these technical problems may be exacerbated by adaptive techniques for monitoring, storage and transmission both from local decisions and those decisions commanded from the larger off-line analysis.
Of course the determination of the appropriate margins for decisions based on the information is key to business opportunity and risk!
___________________________________________________________________________
Let us help you maximise the business potential of your product and its software
Email me on stuart.jobbins@sofintsys.com
Visit our website, or follow us on your preferred Social Media for our latest views. 
___________________________________________________________________________