Knowledge Sharing Impact on Group Problem Solving Productivity

Being that I'm currently working with a product which enables knowledge sharing among a collective group of problem solvers I thought I would develop a model to investigate the impact of knowledge sharing within a group problem solving environment. I actually built this model half a dozen times before I was willing to believe the implications it was providing as the implied impacts were a little too good to be believed. The implications of this model will be conveyed in following sections.

It is my intent to elaborate the development of the model much in the same way that I initially developed it. I find that this makes it easier for others to understand the model rather than having it just thrown at them full force. So here goes.

In a problem solving environment where problems are presented by customers calling on the phone backlog represents the number of problems waiting to be solved. The call rate tends to add to the backlog while the resolution rate tends to subtract from the backlog. If the call rate is grater than the resolution rate the backlog gets larger. If the resolution rate is greater than the call rate then the backlog gets smaller. The question then is what influences the resolution rate?

The resolution rate is a function of the number of analysts and length of time it takes an analyst to solve a problem, i.e. the resolution time. Increasing the number of analysts adds to the resolution rate while the resolution time subtracts from the resolution rate. So next we ask what influences the resolution time.

As analysts solve problems they add to their experience base. An analyst's experience base subtracts from the resolution time associated with solving the same problem the next time it is received. As such, as most environments have some volume of repeat calls, an increase of an analyst's experience base will decrease their overall average resolution time. The extent to which an organization is involved in knowledge sharing has an impact on the growth of this experience base.

When there is no knowledge sharing in effect then each analyst must develop their own knowledge base. If knowledge sharing is being actively pursued it enhances the rate of growth of the each participants knowledge base. This increased growth rate is dependent on the percentage of knowledge sharing in effect and the number of analysts participating. Yet, the act of knowledge sharing effectively adds to the resolution time. This increase in resolution time is based on the fact that it takes time for an analyst to prepare what they have learned and it takes time for other analysts to learn from what is being shared. As such, the increase in resolution time depends on the percentage of knowledge sharing occurring and the number of analysts participating in the knowledge sharing.

There is an additional influence dependent on backlog that should be expressed.

This influence is between backlog and resolution time. The members of an organization have a perception of what is considered an acceptable backlog. As the backlog rises above this level analysts essentially decrease the average resolution time to bring the backlog back in line with what they consider acceptable. This may or may not happen without management attention to the rising backlog. This decrease in resolution time has a limit. As the resolution time decreases analysts begin to experience additional stress and begin to experience burnout. When this happens the average resolution time will begin to decrease. At this point employees will also probably begin to bail out of the organization. This bailout influence is currently not being taken into account.

As the backlog tends to decline below the acceptable level the organization will begin to increase the average resolution time tending to maintain the backlog at the acceptable level. This also has a limit for as analysts become less and less utilized the best analysts will begin to feel underutilized and begin to bail out of the organization. This bailout influence is currently not being taken into account.

The manner in which SolutionBuilder impacts this model should be rather obvious from the fact that it promotes knowledge sharing through an accelerated development of the knowledge base.

Using SolutionBuilder increase the rate of development of the knowledge base. This rate of increase depends on the number of analysts actively engaged in Using SolutionBuilder. Using SolutionBuilder also influences and increase in average resolution time because it takes additional time for an analyst to develop the solution in SolutionBuilder. Note that this increase is substantially less than the increase associated with knowledge sharing as it is not multiplied by the number of analysts for they are not involved with the developed solution until such time as they need to use it. When this usage occurs it shows up as a known problem delivery using SolutionBuilder. A resolution time which is much less than the resolution time associated with the initial development of the solution.

There is one additional influence associated with this model. That is the effect of adding additional analysts to the organization. Since it is seldom that you can add additional analysts and have them be as knowledgeable and productive as seasoned analysts the addition of new analysts essentially reduces the effective number of analysts. This is because experienced analyst time is used to bring the new analyst up to speed. The reduction in effective analysts will decline over time as the new analyst begins to develop experience. It is expected that using SolutionBuilder will both substantially reduce this impact as well as decreasing the time it take for a new analyst to come up to speed.

More to follow:

Download ithink model [, 37k]

theWay of Systems * Feedback * Musings
Copyright © 2004 Gene Bellinger