Problem Formulation

Contents
Problem Formulation
The first step in conducting a significant simulation project is to ensure that
adequate attention has been directed toward understanding what is to be
accomplished by performing the study? These activities are known as the
problem formulation process and consist of:
• A formal problem statement
• Orientation of the system
• Establishment of specific project objectives
A formal problem
statement
A formal problem statement
The problem statement component of the problem formulation step should
provide both the practitioner and the potential audience with a clearly
understandable high-level justification for the simulation.
• Increasing customer satisfaction
• Increasing throughput
• Reducing waste
• Reducing work in progress
1. Increasing Customer Satisfaction
Increasing customer satisfaction is of fundamental interest in any system
involving service operations. This type of system typically includes waiting or
processing queues. Reductions in queue time usually result in increased
customer satisfaction. Customer satisfaction may also involve delivering
products when promised. Reductions in the number of tardy jobs will reduce
operating costs associated with the loss of goodwill and will increase customer
satisfaction.
2. Increasing Throughput
Increasing throughput involves the amount of products or number of jobs that
can be processed over a given period of time. This can involve the elimination or
improvement of different process operations. It can also include the
identification and redesign of bottleneck processes. Reducing waste results in
reduced operating costs and increased net profits.
3. Reducing Waste
Waste can be reduced through reductions in spoilage and obsolescence.
Spoilage can involve processes that are time and temperature critical.
Obsolescence waste can result from an organization’s inability to bring its
product to the market on time.
4. Reducing Work in Progress
Work in progress is work that requires further processing for completion. Work
in progress is commonly found in processes that require multiple discrete
operations. Work in progress typically requires storage before the next process
can be carried out. Reducing work in progress reduces process costs associated
with resource capacity and storage requirements. Large amounts of work in
progress can result from insufficient resource capacity or poor operating
policies. Reducing work in progress can decrease the space needed for
manufacturing or distribution facilities.
5. Tools for Developing the Problem
Statement
There are two common tools available to the practitioner for assisting with the
problem statement. These are not actually problem-solving tools but rather are
problem identification tools. Although these tools were originally developed for
the manufacturing environment, they can easily be adapted to other sectors of
industry. These tools are:
• Fishbone chart
• Pareto chart
Fishbone chart
The Fishbone chart is also known as the cause-and-effect diagram, man–
machine–material chart, and as the Isikawa chart (Suzaki, 1987). The purpose of
this chart is to identify the cause of the problem or effect of interest. The
Fishbone chart looks similar to the bones of a fish. The head of the fish is labeled
with the problem or effect. Each major bone coming out of the spine is a
possible source or cause of the problem. For example, in a manufacturing
process there are major bones for man, machine, material, and methods.
Fishbone Chart
Pareto Chart
The Pareto chart is a second technique to help the practitioner to develop the problem
statement (Nahmias, 1987). It may turn out that there are several sources or causes of the
problem or problems of interest. The concept of the Pareto chart was developed by the Italian
economist Wilfredo Pareto, who postulated that only a few factors are the cause of many
problems. This is frequently referred to as the 80–20 rule. The 80–20 rule states 80% of the
problem is caused by 20% of the factors. The 80 and the 20 are not magic numbers but seem to
be approximately observed in many instances.
To implement a Pareto chart, you would need to know the number of defects from each source
and the cost associated with the defect. If these are multiplied together, the true cost of each
defect can be determined. It is possible to end up with four general combinations of true costs:
1. Large number of inexpensive, easily repairable defects 2. Large number of expensive defects
or a no repairable product 3. Small number of inexpensive, easily repairable defects 4. Small
number of expensive defects or a no repairable product.
Pareto Chart
The Pareto chart would clearly indicate the amount of money lost to the large
number of expensive defects or nonrepairable products, which should be the
focus of the problem statement. The next question would be whether the small
number of expensive defects or nonrepairable product cost is greater than the
large number of inexpensive easily repairable defect cost. Last, small numbers of
inexpensive easily repairable defects would be the subject for a later study. By
understanding the true cost associated with the process, the simulation
practitioner can focus his or her limited time on those few factors that would be
expected to have the maximum benefit for the organization.
Orientation
Orientation of the system
Orientation involves the practitioner’s familiarizing himself or herself with the system. In most
cases the practitioner will not have the depth of knowledge of the managers, technicians, or
workers who are involved in the process on a daily basis.
If the practitioner has had some contact with the system, the practitioner may feel as though he
or she already has sufficient knowledge to bypass a formal orientation. Even if this is the case,
abandoning the system orientation is likely to be a poor decision.
The reason for this is that even though the practitioner may have general familiarity with the
system, it is unlikely that the practitioner is familiar with the system from a simulation
viewpoint. The practitioner must revisit the system keeping in mind how the system should be
modeled and analyzed with respect to the nuances associated with simulation modeling.
Orientation Process
The orientation process should consist of a minimum of three different types of orientation
visits. We recommend that the visits be conducted separately to increase the probability of
understanding the operation of the system. However, in extreme cases, there is no operational
reason why these visits cannot be performed during the same time period as long as the visit
requirements can be met.
The three different types of orientation visits are:
1. Initial orientation visit
2. Detailed flow orientation visit
3. Review orientation visit
Initial Orientation Visit
The purpose of the initial orientation visit is to obtain a high-level understanding of the basic
inputs and outputs of the system. This type of orientation is typically conducted as a guided tour
by a member of the commissioning organization.
It is important not to attempt to understand too much detail of the system in the initial visit. If
the guided tour lasts longer than a couple of hours, it is likely that too much information is being
presented to be digested at this point in the orientation process.
After the initial orientation visit, the practitioner should immediately reflect on the major
components of the process while recollection of the visit is still fresh. The practitioner should
identify particularly complicated components.
Detailed Flow Orientation Visit of the
system.
In the first detailed flow orientation visit, the practitioner can request additional clarification on those
components. After that he gather detailed information on the operation of the system. If the system
is complicated or contains a number of components, the practitioner may need to conduct multiple
visits.
The practitioner should make detailed notes on the system operation by recording:
• The types of entities that are processed by the system
• The number and type of queues
• The number and type of system resources
• The sequence of processes as experienced by the entities
• How the system performance can be measured
After the detailed flow orientation visits, the practitioner should begin developing an understanding
of how the system operates. It may be beneficial to begin the development of the high-level flow
chart ultimately required for the system definition phase of the project.
Review Orientation Visit
Once the practitioner has an understanding of the system operation, it is time to return to the
system for the review orientation visit.
The purpose of this visit is to ensure that the understanding of the system operation is
consistent with the practitioners’ understanding of the system and/or flow chart. It is also not
unusual for the practitioner to observe new aspects of the system even after the review
orientation visit has been conducted.
During the review orientation visit, it would be helpful to reengage the services of the system
representative to answer any last minute questions and to insure that the practitioner’s
understanding of the system is correct.
Orientation Example
Consider the relatively simple process of airline passengers taking their seats in a commercial jetliner.
Close observation of this process will indicate that we will need to have knowledge of the following
items:
• The number of passengers on the plane
• The type of passenger: regular, family, handicapped
• The rate at which passengers travel down the aisle
• How far back the passenger must travel down the aisle
• Whether the passenger has an aisle, center, or window seat
• The percentage of passengers with carry-on bags
• The time it takes passengers to load each piece of luggage in the overhead bins
It is obvious from this list of observations that the seemingly simple process of taking a seat in an
aircraft suddenly becomes more complex when viewed from a simulation modeling standpoint.
Tools for the Practitioner’s Orientation
The practitioner should have
• a high degree of access to the system.
• A high degree of access may enable the practitioner to employ electronic recording devices
such as a digital camera or camcorder.
Project Objectives
Project objectives
This is the final step in the problem statement process of the simulation study. When the project
objectives are being established, it should be obvious that the practitioner develops the
objectives in close cooperation with the individuals commissioning the simulation study. Though
this may seem obvious, any breakdown in communication during this critical phase can have a
devastating impact on the value of the simulation to the commissioning organization.
The project objectives should also be considered as dynamic. During the course of the study, the
practitioner may discover other information that may prove to be more profitable to pursue.
Sometimes the commissioning organization does not initially have as much interest as would be
beneficial. This means that the practitioner, who may or may not have sufficient domain
knowledge of the system, is forced to suggest the initial project objectives. Only after the
credibility of the practitioner and simulation modeling process has been established does the
necessary level of interest appear. If this does occur, one hopes it will not be too late to adjust
the simulation project objectives.
Examples of Project Objectives
The following list of common project objectives can be used by the practitioner as a starting
point to develop a project-specific set of objectives. Common project objectives may involve
• Performance-related operating policies
• Performance-related resources policies
• Cost-related resource policies
• Equipment capabilities evaluation
Decision-Making Tools for Determining
Project Objectives
These tools can include (Gibson et al., 2003):
• Brainstorming
• Nominal group technique
•Delphi process
Brainstorming
The idea behind brainstorming is to generate a storm of ideas that are created by being exposed
to other ideas. In a brainstorming session, participants will often come up with ideas that
otherwise they would not have. Discussion of other ideas gives birth to new ideas. A
fundamental aspect of brainstorming is that no participant should criticize the validity of
another participant’s ideas. If any criticism exists, some participants may feel uncomfortable and
be reluctant to share openly their ideas.
Brainstorming Process
Electronic Brainstorming
There are a number of advantages to the use of electronic brainstorming over conventional
brainstorming. These include the ability to include more individuals in the discussion and the
participation of individuals who could not attend the session because of distance or schedule
limitations.
It also allows for an electronically recorded history of the discussions. The primary disadvantage
is the unsynchronized nature of implementing electronic brainstorming. Unless everyone in the
session is simultaneously participating, the discussion does not occur in real time. Thus, it may
be necessary to allow for several days before the session can be concluded. Electronic
brainstorming can be implemented in either of the following ways:
• E-mail discussion list
• Electronic bulletin board
Email discussion list
An e-mail discussion list is the easier of the two electronic brainstorming methods to
implement. With this method, the practitioner sets up an e-mail address list that includes the e-
mail addresses of all of the individuals with potential interest in the project. These individuals
are commonly known as stakeholders. The practitioner is directed to Chapter 3, “Project
Planning,” for an in-depth discussion of the different types of stakeholders. As each individual
thinks of a new idea, the concept is entered in an email reply to the list and is automatically
redistributed to all of the members of the e-mail list. A small irritation that has been identified
with these types of e-mail list brainstorming sessions is when the participants set their e-mail
package to notify them automatically of incoming e-mail. If there is high volume on the list, the
constant notification of e-mail messages can become disturbing. Many otherwise useful
individuals are more likely to request to be removed from the list rather than be continuously
disturbed by the series of e-mail messages.
The electronic bulletin board
The electronic bulletin board approach to brainstorming eliminates the continuous e-mail
message notification problem. Here, the practitioner sets up a Web-based bulletin board to
control the brainstorming discussion. Individuals must take the initiative to regularly check the
bulletin board for new developments. The potential success of this approach may at first seem
questionable. However, many user group bulletin boards typically experience high levels of
activity. This activity includes a number of topic professionals who are completely willing to give
expert technical advice free to complete strangers. If the practitioner’s brainstorming group is as
interested in the project as the average individual is in interest group bulletin boards, the
necessary level of activity is assured.
Nominal group technique
Once the storm of ideas for the project objectives has been generated, it is the administrator’s
job to determine the most important project objectives. These objectives can be identified by
using the “nominal group technique.” In this procedure all of the participants have the
opportunity to vote on what they believe is the most important project objective. All of the
participants are given an equal number of votes to cast. Each vote holds the same weight
regardless of the status of the voter. This equality will encourage the less senior individuals in
the group to vote without undue influence. Before the nominal group technique process begins,
participants may ask for clarification on different project objectives. Should this occur, the
parent of the idea should provide a brief response. Once the process is ready to start, the voters
are asked to cast one vote per cycle for an individual project objective idea. This cycle repeats
until all the votes are cast. Voters may vote for the same idea until all of their votes are cast, or
they may split their votes among different ideas. At the end of the nominal group technique
process, the administrator should compile the votes and identify the most popular project
objective ideas. These project objectives will then guide the rest of the project activities.
Nominal group technique
Delphi process
In some organizations or situations undue political influence interferes with the selection of the
project objectives. If it is desirable to minimize the effect of these influences, the administrator
may elect to utilize the Delphi process technique. This process serves the same function as the
nominal group technique; however, all voting is conducted anonymously after the brainstorming
session has concluded. The process begins by having the administrator distribute a list of all of
the brainstorming ideas. The participants then cast one vote for a particular idea. At the end of
the first cycle, the administrator retains only a set of the most popular ideas. The new set of
ideas is redistributed to the group. This process continues until the set of project objective ideas
is a suitable number. The number of ideas to advance to the next round is at the discretion of
the administrator. One method is to eliminate any ideas that receive no more than one vote.
This means that that idea is of predominant interest to only one individual. As with the nominal
group technique, the final list of project objectives guides the rest of the project activities.
Delphi Process