Problem Definition

The first step in the System Improvement Process (SIP) is to define the problem. The guiding principle of this step is: The more correctly a problem is defined the less work required to solve it. Difficult complex system problems are best defined using this standard format:

Move system A under constraints B to goal state C by deadline D with confidence level E.

The letters are the variables that are filled in to define a problem. Variables C, D, and E are the solution goal, the most important part. Constraints B include human resources, budget, conflicting goals, uncertainty, authority limitations, and so on. The problem is solved when a solution is created that will move the system to goal state C by deadline D with confidence level E.

System A

Discussion on defining the system with the problem quickly led to the notion that it is probably too late to solve the Port of Hastings problem to the full satisfaction of all parties. Better is to see the POH problem as more evidence of a recurring defect in a planning process that is broken. As Glenn wrote in a pre-meeting email:

"The problem from my perspective is the process used to come up with the draft. The groups involved all seem to have a vested interest to encourage development. It is as if no one has stopped and asked the question do we need this? The focus of the process to establish the strategy is all around increased infrastructure and development. Any reference to social or environmental needs seem to be more down the lines of mitigating issues as much as possible rather than preventing or avoiding."

This perspective allows us to see that the real problem is the present planning process produces too many defects, in the form of bad decisions. The systems that are doing this are small units of government, like the Port of Hastings Corporation. These provide public infrastructure, like ports, road systems, and electrical power. All use a similar social control model, which is a collection of rules describing how a social unit works.

This leads to our definition for System A. It is the social control model used for small units of Australian government that provide public infrastructure. Within this system we will be focusing on the process it uses to make decisions affecting the common good of the people.

Constraints B

Constraints are limits imposed on solution effort and actions, such as laws, budgets, conflicting goals, uncertainty, authority limitations, etc. The System Improvement Process treats all constraints as part of the problem to be solved.

Several constraints have been identified. More will come. So for we have:

1. A very small minority of citizens are questioning the decisions of units of government like the PoHC.

2. Lack of needed information - Some citizens are opposed to certain decisions. But most appear to want additional information, to convince themselves that the decisions are fair and well founded. Usually the lack of information is caused by it has not been created, such as an environmental impact study or a detailed demand for services analysis. Sometimes the information already exists. Then it can be provided. However, sometimes it exists and is withheld, and kept secret.

3. Secrecy - To often units of government keep important information secret. For example, in the Port of Hastings problem, some citizens see the signs that the proposed rail system is so large that it is secretly being designed to also support nuclear fuel and waste transport. The plans for the port were in some cases apparently kept secret for a year, and then suddenly released on an unprepared public. This has contributed to an adversarial relationship, which would be greatly reduced if planning was transparent.

Goal State C

The goal state is the state the system should be in to consider the problem solved. A goal state must be quantified, so that progress toward it is measurable, and not vague and subject to interpretation.

The system is the social control model used for small units of Australian government that provide public infrastructure. It is not the units themselves, though our discussions frequently drift into seeing it that way for simplicity.

The goal state of the social control model is that we will have tangible proof that if adopted, it would cause those units of government using it to optimize the common good of the people affected over 80% of the time.

This forces what the common good is to be defined. Presently this is not being done. How this can be done will be investigated as this project proceeds. For example, maybe specific attributes of quality of life will be measured. Maybe polls will be taken about how populations feel about how well their common good has been achieved.

This is such as ambitious goal state that our goal will not be to achieve all of it, but to make such significant progress toward it that later projects can take it the rest of the way. Exactly how to measure that significant progress is to be defined.

Deadline D

The significant progress will be achieved in 12 to 18 months, which is also the like span of this project. If we can demonstrate in one typical case that we have corrected whatever is causing the present defects, then we have proven that we have a new social control model that works. It can be then be replicated. Thus we do not have to have corrected the behavior all units of government by this deadline, which would be way too ambitious.

Confidence Level E

Our work mill be managed such that we have a 90% probability of achieving the goal state by the deadline. A higher probability would require a large, well funded project and the cooperation of the Australian government, which we do not have at this stage. A lower probability would mean that failure is okay and we don't need to do our best, which is not the case. The 90% is a ballpark estimate. It allows us avoid the trap of no confidence level, which would make it impossible for anyone to rationally decide how much of their time and money to invest in this project.

Summary

We define this problem as how to move the social control model used by small units of Australian government for public infrastructure to where the model will work at least 80% of the time to optimize the common good of the people affected. Our deadline is 12 to 18 months, with a confidence level of 90% that we can do this.

Yes, this will be difficult. Yes, amazing breakthroughs will be required. But breakthroughs do not come unless you have defined something to break through.