Monday, March 9, 2015

Macro vs. Micro Level of Systems Analysis

In general, the qualitative or constraints come from the macro level; the quantitative or energy come from the micro level. 

The organization's become an over-complex and hyper-connected digital ecosystem nowadays, analytics turns to be a “must have,” not a “nice to have” tool, for both strategy making and problem solving. Macro vs. Micro level of systems analysis, what are they all about, and how to apply them accordingly?


Macro analysis identifies the impact factors in a system, and micro-approach is pinning down into the underlying elements: Macro-approach identifies the impact factors in a system into chunky distinct facets or scenarios. It can be articulated via Alexander's "pattern language," where the "patterns" are essentially relational process or entity templates with attached descriptions and contextual information that make them useful in a range of cognate cases within a target domain of choice. On the other hand, the "micro-approach" is more concerned with pinning down the elementary and underlying constituents of a system, seeking to abstract and codify aspects of the system into the data-sets and algorithmic process formulas. In such scenario, the individual and elementary constituents of a resultant "system object" (or composite of objects) may be meaningless in isolation, unlike at the macro-level that concerns itself are primarily with the larger "macroscopic objects" of the system.


The diamond-shaped macro and micro perspectives are both analyzed and synthesized to establish a direction, but these are iterative. Where macro or micro are simply methods of viewing or understanding resolution of the process. Analysis and synthesis is iterative, first to ask, then to answer, and again, throughout the process in refining the ideas. You could also recognize the entire method as "macro" and the iterations as "micro." If the system structure is a diamond made up of two triangles; The top triangular structure of the diamond as Macro Analysis and the bottom triangle of the diamond as Macro Synthesis. The analysis and synthesis as articulated throughout the system process of refinement; first via divergent thinking then via convergent thinking are micro iterations of the system. One could imagine the system map as a diamond shape, first diverging to articulate the various factors as defined by the system, then converging in a process of articulating solutions; in tighter and tighter iterations of analysis and synthesis to refine and solve for the system, articulated at the bottom point as a holistic set of ideas that refined converge to a systemic solution.
Macro:
- identify system/subsystem input/output boundaries for a system, as well as external variables to the system that can and cannot be controlled.
- identify the subsystem interactions
- data collection on subsystem input/output
- modeling and analysis of the system/subsystems
Micro:
- identify subsystem internal boundaries and interactions, processes, flows, etc.
- data collection
- modeling and analysis of the subsystem


Macro analytics is more qualitative; and micro analytics is more quantitative. Macro analytics model is often the starting point when one is ignorant about the details - more qualitative and general, and "micro" analytics models are "scientific" and specific - more quantitative. The interesting observation is that as one gets more "micro,"  more data is required to set parameters and the model has less generality. And this observation applies equally to all length scales - from cosmological to cellular to atomic. It starts as a top-down method of divergent thinking to establish context of the system that will be eventually conceptualized and designed for. This obviously requires both levels of macro and micro perspectives of this system. The process uses iterative methods of both analysis and synthesis to develop the system structure and divergent framework. At the bottom of this framework the method elicits both functions of the user and functions of the system. Then data is cross-referenced, also this divergent map provides the context to converge, toward the solution space. Evidently, the macro-level is more tangible, intuitively meaningful and immediately actionable -- a precision level of choice for illustration and management. Yet it doesn't yield itself well to the generation of metrics data on the process progression, to logic validation, or to other forms of programmatic evaluation. Analysis with a deeper itemized reach would confer obvious benefits in terms of identifying process fallacies, providing spectrums of perspective, or even rendering (potential) novel solutions based on the available data.


Divergent-convergent thinking cycle:  Where many 'designers' utilize the funnel to parse ideas of the 'fuzzy-front-end,' from brainstorms, but by starting with convergent thinking, only funnels thoughts swirling into a muddy bucket of thought. However, if one starts by diverging instead, like turning a tree upside down, starting at the trunk on top and then gradually building finer detail to the tips of the branches. This upside down tree establishes a rich context, focusing on coverage and breath that establishes a meta-understanding of the system being designed for. Then one can place the funnel underneath, to vet and converge elements into usable ideas and thoughts. Holistically this system structure illustrates a plan, where the system context is clear, the concepts are clear and the solution set will be clear. The part of the analysis, where both self-properties and relational properties are identified, and then proceeding to analysis of the processes where the identified entities are involved, again both self-processes and relational processes, or combinations thereof.


It is an "artificial boundary" design requires both analysis and synthesis: to ask and to answer, the only constant is change, especially if you are considering the dynamics of the system. First, discover the issue, analyze. Next, frame the issue, synthesize. They look at all these factors through different resolutions of macro and micro 'lenses' or methods. And then build concepts based on these resolutions. This builds a structure used to plan the process, and  it's an "artificial boundary or constraint." This is the very definition of "synthetic" modeling:
-Macro view is mainly to know what to do,
-Micro view is mainly to know how to do it,
-Both micro and macro requires operational methods: Analysis, Synthesis.


Systems thinking provides a perspective that characterize, synthesize, and thus recognize both patterns and anomalies of a contextualized system. Fortunately or unfortunately, humans are the major outlier of any system. As many data-points we can track we can not 'know' every aspect of natural systems. This is why it is critical to recognize the perspective as taken, when articulating 'systems thinking.' Macro and micro refers to the level of aggregation, both can be collaborative, analytical, synthetical, competitive, blue, yellow and so on. Using a micro-meso-macro approach. The meso level is the customer experience. The micro level is the passion and talent of the designers.The Macro level are the constraints of time and money. A small change in the macro level is most effective to create a big change in the meso level.The macro view is useful to tell you What? While micro view tell you How? You have to be careful not to see this as a view, because it is up to you. Both are necessary, analysis and synthesis. Continue to questioning: Why?, Where? and Who? Those three friends require their own views like fishbone, mapping and agents. All of them are necessary to play the imitation game.


As "everything exists in time," all systems can be related to the system model at four levels. This articulates a dynamic system structure. From here one can imagine user needs and system requirements that articulate and solve for the established context framework, allow participants to bring up their own problems, and then see how solutions to complex problems become self-evident through the power of paying attention. Under such collaborative and competitive systems, there’s conflict and crisis. As such the term 'environment' doesn't mean anything as a better lens for an ‘'ecosystem', as a collection of overlapping changing systems interacting at and across as different levels of abstraction over time. As such systems can be resolved from multiple perspectives in time and resolutions change with time. Underlying Change in those ecosystems isn't just aggregate stress cause-effect relationships from interaction but could be by chance. The four levels of system analysis model are:
- Top level defines the system scope,
- Second level describes the modes or major operations of the system,
- Third level defines the activities of the modes, and the
- Fourth level describes the user and system functions of the activities.


While oscillating between two levels of system analysis is important, it may also be useful to keep in mind that the macro system has more "dimensions of analysis" that should be considered. For example, on a micro level, one department may improve their operational efficiency. On a macro level, you should consider the optimization of all the departments, and the effectiveness and efficiency of the inter-departmental interactions (an additional dimension). Indeed, the macro may be said to be defined by those interactions. Sometimes, learning by doing - working with dynamic change as your tool is the best approach, rather than analyzing, planning then implementing, and waiting. A basic skill is determining which approaches are appropriate, rather than just always reaching for system thinking first. In general the energy comes from the micro level. The constraints come from the macro level. The boundary definition of the "problem" is at the meso level. Analogically, when you drive a car, the GPS gave you the macro view, while you are engage in the micro view. Business is a complex and adaptive system and you are on the wheel. You can put in practice your strategies and tactical moves, when you navigate through the journey you’ve never been before.


0 comments:

Post a Comment

Twitter Delicious Facebook Digg Stumbleupon Favorites More