As currently available, nearly all process platforms are organized for central design of the process and, by extension, require central control of the process. This sort of systems design does a poor job of identifying and enabling process improvement.
The best central design process platforms are spreading out control within a hierarchical system that is designed for central control. Sometimes these systems can succeed, as in the case of the Toyota manufacturing system or Best Buy's ROWE system. But they seem to be working against the grain. Existing platforms work against the management maxim to push decision making as far down the hierarchy as possible because they demand that the final decision be made in a central group, or even more awkwardly, by a consensus of central groups that have grown up their own bases of power.
This is in contrast to a mythical system that allows for organic process growth by facilitating small (or even large) process changes at the periphery, without any involvement from a central decision-making organ. The work of the organization is then to measure the effectiveness of these slightly changed processes in the real world rather than in a mental model or simulation. Decision making becomes a matter of choosing the right measures for these processes, itself a process that can be measured. The measures go all the way up the chain to "what do we want to accomplish?" and "what are the constraints on how we ought to get there?"
At what level should processes be personalized? How far down the chain should the decisions be pushed? Ideally, any person who has a job description be able to control their job process, at least within limits. An important part of this is giving each person the resources to understand and evaluate the impacts of changes to their processes. These impacts might be in productivity, cost, environment, employee/customer satisfaction, etc. Impacts are also on multiple levels: the individual, the department, the group, the division, the company, the industry, etc. The resources for understanding and evaluation will inevitably be focused on a few of these facets but should acknowledge the existence of all of them.
What about the risk? How can we allow people to run off and change the way we do business without vetting the impact before-hand? They might mess something up.
I'm pretty certain that we're already messing up. There's definitely a lot of upside potential in the organizational effectiveness department. The keys here are going to be developing organizational strategies for understanding and evaluation rather than developing securities and controls, as most damage that is done by errant employees appears to be done through a lack of understanding rather than any sort of maliciousness. The effectiveness of most IT and process controls is in their ability to prevent people from making inadvertent mistakes rather than knowingly causing damage. Once this is accepted it becomes clear that establishing control is secondary to establishing understanding.
Organizations may begin to focus the design of their process systems on aligning change and impact, thereby allowing for isolated changes that avoid side-effects across the organization. A sort of functional programming paradigm of process design. This is not the only option, but it fits in with the ongoing theme of decoupling business units. Another option might be to control change to be within certain predetermined guardrails, or to be in accordance with business rules developed in a parallel business rules and controls system, to allow for maximum optimization and flexibility within a more integrated business.
I hesitate to say that this will enable "evolutionary process", as it seems unlikely that we will never achieve the level of iteration and competition that might be necessary for evolutionary process design in the near future. But there appears to be a lot of room for tapping individual and small-group initiative both to improve processes at an organizational level that is not currently seen as cost-effective, and to relieve frustration with centrally developed and decreed processes.