> Deployment in CAEBM technology is rather straight forward: Trained and tested EBMs can be fed into packages of code in contemporary programming languages. This makes the contained know-how independent from any sources and highly portable, features which are not always welcomed by local know-how sources, especially in traditional organizations, or by cautious problem owners.

> In continuous KHE environments, the EBMs can be updated continuously, providing a kind of adaptive modeling, if the problem owner wants such thing. 

> In time-critical applications the code packages can be optimized appropriately, or the neural networks even can be used to establish (natural) high-dimendional parallel computing, opening the chances for exceptional high-speed real-time applications.

> Because of the high portability of know-how, due to CAEBM, rather new business models, eg in provider - customer relationships can become reality: For example, a welding machine manufacturer may collect continuously the example-based application know-how of his customers, concentrate it, and distribute it to new (inexperienced) customers, together with a new welding machine as an add-on. Also old customers can participate in this win-win arrangement. 

> Even special know-how concentrated businesses become possible, which collect or buy (continuously) know-how from different sources, concentrate it, and sell it to interested organizations. Honoring it's well-recognized importance, know-how, in general, can become an important product of its own, thanks to CAEBM.

> So, over all, also at deployment time the CA aspect of CAEBM technology can make a most important difference.