Whenever and wherever we have a system we have a potential mess problem. Systems are designed taking each single participant or entity into context, finding the relationships with other entities, finding the optimum path etc. This is a rather constricted way of designing due to which it’s difficult to foresee a mess coming up. When the entire system is integrated and put to uses then we realize what mess problems crop up.
Again for solving these mess problems people break up the system into parts and try to solve the problem. This is a faulty approach for solving mess-related issues because:
v Unless we see the system and the mess in it’s entirety we run the risk of not solving all the problems fully.
v Further systems are dynamic so unless a robust solution is found out we can have mess develop after further use of the system.
Examples of messy networks include traffic networks, internet and LAN networks, switching controls etc.
Below is a diagrammatic representation of a mess in a typical computing network
.
(Source: www.acm.pku.edu.cn)
Messes are complex and dynamic systems and each mess shows these characteristics:
1- Merge Point: Point where 1 or more entities merge.
2- Burst Point: Point of emergence of 1 or more entities.
3- Loops: Loops are inter-connected entities affecting each other. Loops are again of 2 types: Positive loops are those where actions in one entity cause an action in a similar direction in another entity. Negative loops are those whose actions cause a change in an opposite direction.
A system drawn as above with loops and points is called an influence diagram (ID). A typical ID looks as below.
An ID contains arrows and nodes which are of different shapes and convey different meanings.
(Source: www.lumina.com)
The 1st step of avoiding a mess is to draw an Influence Diagram for the system. Then with the help of reference projections we can show what the future of that system will look like. What this implies is that the system is projected into the future and the mess revealed. This method is based on two faulty assumptions as follows:
a- System will not change its current configuration.
b- System will change only as expected.
This will help in knowing where we are going to in the future and hence we can avoid any messy situations. Further this method opens our eyes towards the formation of mess which would have gone un-noticed had we not been able to see the future condition of the system. So knowing the future will help in avoiding the mess.
Finally to conclude mess management requires lot of creativity and planning. All methods to manage a mess ultimately depend on the planner’s ability to perceive and foresee the possible messy situations and act accordingly.
Reference:
Ackoff, R.L.(1986). Management in Small Doses, Wiley, NY.
Network Mess. http://www.acm.pku.edu.cn
Influence Diagrams. http://www.lumina.com
(The work above is of my independent doing and all sources have been duly acknowledged.)
No comments:
Post a Comment