Consistency and economies of scale constitute most pivotal features of processes. They facilitate operations that run efficiently day in and day out with predictable outcome and uniform quality. Paradoxically, in today’s service economy high-grade process delivery requires the consideration of variation in customers, environment, used resources, scenarios and various other aspects. It is thus often inappropriate to directly apply traditional process control and process management tools used, for example, in mass production of products, where standardization is prevailing and variation is minimized. In order to still address key management issues such as quality, cost and schedules in service processes and production processes of customized products alike a proper characterizatio...[ Read more ]

Consistency and economies of scale constitute most pivotal features of processes. They facilitate operations that run efficiently day in and day out with predictable outcome and uniform quality. Paradoxically, in today’s service economy high-grade process delivery requires the consideration of variation in customers, environment, used resources, scenarios and various other aspects. It is thus often inappropriate to directly apply traditional process control and process management tools used, for example, in mass production of products, where standardization is prevailing and variation is minimized. In order to still address key management issues such as quality, cost and schedules in service processes and production processes of customized products alike a proper characterization of variation in processes is imperative.

To date, however, there is no attempt in trying to systematically explain and statistically assess the characteristics of dynamic process implementations in a holistic manner, i.e. starting from the initial need for variation to the final actual implementation in the process. This research therefore formalizes the concept of process context as the integral source for process variation under the hypothesis of an existing causal relationship with the process. At the same time, it is theorized that a rising variation in this process context not necessarily causes an increase in process variation, thus creating the opportunity for various process realization strategies. For modeling the process context variation and its translation into dynamic process behavior approaches found in two discovered independent analogies in the field of theoretical physics and information theory are adopted. A practical implementation is proposed and applied in two relevant real cases with extensive data sets. With the dawning digital age such fine-granular process data is now abundant.

As a result, a methodology has been found for the first time that allows for holistically explaining, modeling and measuring variation in processes. It introduces the fundamental concept of entropy to the process field and manifests a theoretical understanding of possible response mechanisms underlying the wide range of detached process realization strategies. In terms of an analytical framework it facilitates a data-driven characterization of processes while tackling the need for an estimation approach and the impact of two involved trade-offs. Both case studies have shown the fitness of the modeling and corroborate the underlying hypotheses. The novel concept of responsive processes hence establishes as a central contribution. It is defined on the basis of newly derived fundamental entropy measures. Its reinterpretation of variation in processes as a mandatory or even desirable feature turns an accomplished management of a responsive process into a remarkable competitive edge. This relaxation of a persistent focus on standardization has further been found to pave the way towards a novel consistency paradigm. Building on this consistency requirement a hypothesized economic limit for process entropy has been proven.