DECISION SUPPORT SYSTEM
ARIZONA STATE UNIVERSITY
CIS 235 - INTRODUCTION TO INFORMATION SYSTEMS
WEEK 6
8.1.
Definition of Complaint:
The reason customers complain in general is because they feel dissatisfied with the
services provided, resulting in customers demanding dissatisfaction with the services
provided. As previously quoted that service is a very important aspect to provide, if the
service is good, the customer will appreciate and feel satisfied with the service provided or
vice versa, if the service is bad, it will end in protest (dissatisfaction) or what is referred to as
a complaint.
According to Tjiptono in (Darmajaya: 2016) argues that complaints or complaints can
be interpreted as an expression or sense of disappointment. Organizations can collect
customer complaints in a number of ways, including suggestion boxes, customer complaint
forms, special telephone lines, websites, comment cards, customer satisfaction surveys and
customer exit surveys. This situation is known as the "Recovery Paradox".
According to Daryanto and Setyabudi, (2014: 32) "Complaints or complaints are
complaints or submissions of dissatisfaction, discomfort, irritation, and anger over services or
products".
This is also stated by Bell and Luddington (2016: 78), that "Customer complaints are
feedback from customers addressed to companies that tend to be negative. This feedback can
be done in writing or verbally".
Based on this understanding, the definition of a complaint or complaint is a service
complaint or complaint is an expression of feelings of dissatisfaction with service standards,
actions or inactions of service providers that affect customers.
8.2.
Problems Complained by Customers:
According to Bell and Luddington (2016: 98), customer complaints are usually due to
problems such as lack of responsibility (responsiveness), lack of help from company staff
(helpfulness), product availability (product availability), store policy, and service recovery.
8.3.
Customer Complaint Program:
Based on various consumer psychology research, complaints can be divided into two
types instrumental complaints and non-instrumental complaints. Instrumental complaints are
complaints expressed with the aim of changing an undesirable situation or circumstance.
Complaining customers can be a gold mine of success but also the beginning of future doom.
According to Leboeuf in (Darmajaya, 2016) there are three main reasons why customer
complaint programs provide great benefits:
1.
Complaints reveal areas that require improvement.
2.
Complaints are a second chance to provide service and satisfaction to disappointed
customers.
3.
Complaints are an opportunity to strengthen customer loyalty.
8.4.
Aspects of Handling Customer Complaints:
According to Tjiptono (2017: 351) there are four aspects of handling customer complaints or
complaints, namely:
1.
Empathy towards angry customers In dealing with emotional or angry customers,
customer service staff must be level-headed and empathetic. If not, the situation will get
worse. For this reason, it is necessary to take the time to listen to their complaints and
try to understand the situation felt by the customer. Thus, the problem at hand can
become clear, so that an optimal solution can be pursued together.
2.
Speed in handling complaints Speed is very important in handling complaints. If
customer complaints are not dealt with immediately, then the dissatisfaction with the
company will become permanent and irreversible which will cause the company's
image to look bad and can spread to the wider community. Meanwhile, if complaints
can be handled quickly, then the possibility of customers the customer is satisfied. If the
customer is satisfied with the way the complaint is handled, then it is likely that he will
become a customer of the company again. The results of the Technical Assistance
Research Program research (cited in Nauman and Giel) show that:
a. 70% to 90% of customers who complain will do business again with the same
company if they are satisfied with the way complaints have been handled.
b. 20% to 70% of customers who are not satisfied with the way complaints are
handled will not do business with the same company.
c. Only 10% to 30% of customers who have a problem (but do not complain or ask
for help) will do business with the same company.
3.
Obligations or fairness in resolving Company problems or complaints must pay
attention to aspects of fairness in terms of cost and long-term performance. The
expected results are of course a 'win-win' situation (fair, realistic, and proportional)
where both the customer and the company benefit equally
4.
Ease for consumers to contact the company
Consumer access to the company in order to submit comments, suggestions, criticisms,
questions, or complaints is a crucial factor that must be considered carefully. Here it is
very necessary to have an easy and relatively inexpensive mode of communication,
where customers can submit their complaints. If necessary and possible, the company
can convey a toll-free telephone line
8.5.
Benefits of Effective Grievance Handling:
According to Tjiptono (2017: 349) there are benefits if we handle complaints effectively,
namely:
1.
The service provider gets another chance to repair its relationship with the disappointed
customer.
2.
Service providers can avoid negative publicity.
3.
Service providers can understand the aspects of service that need to be improved in
order to satisfy customers.
4.
Service providers are able to identify and follow-up on
5.
Employees can be motivated to provide better quality services.
The process of handling complaints effectively starts from identifying and
determining the source of the problem that causes customers to feel dissatisfied and complain.
The source of the problem needs to be addressed, followed up, and strived for so that in the
future the same problem does not arise. In this step, speed and accuracy of handling are
crucial. Dissatisfaction can be even greater if complaining customers feel that their
complaints are not resolved properly. This condition can cause them to be prejudiced and
hurt. The most important thing for customers is that the company should show attention,
concern.
Regret the disappointment of the customer and try to improve the situation. Therefore,
the employees of the company, especially those on the line The frontline needs to be trained
and empowered to make decisions in order to handle such situations. In addition, top
management involvement in handling customer complaints also has a positive impact. This is
because customers prefer to deal with people who have the power to take decisions and
actions to solve their problems. Moreover, customers will feel that the company pays great
attention to each of its customers' problems, and is always trying to improve.
8.6.
Purpose of Customer Complaint
In Tjiptono (2017: 101) in essence there are main purposes for customers to complain:
1.
To cover economic losses, which are usually realized by conducting a voice response
or third party response, this response is aimed at external objects that are not directly
involved in the unsatisfactory experience (for example, newspapers, consumer
organizations, legal aid organizations, and so on). The form of response can be in the
form of claiming legal compensation, complaining through the mass media. (letters to
the editor, newspapers, etc.) or directly to consumer organizations or legal agencies.
Such actions are feared by most companies that do not provide good service to their
customers or companies that do not have effective complaint handling procedures.
2.
Improve self-image (Self Image), if the customer's self-image is closely related to the
purchase of certain goods and services, then dissatisfaction with certain purchased
goods or services.
8.7.
Principles of a Good and Effective Complaint System
According to Tjiptono (2017: 351) quality management is a management system that
implements a good and effective complaint system based on the following six principles:
1.
Visibility
Own and set up complaint channels (CS dept or other customer-facing departments,
phone or call center, email, web, suggestion box) so that customers can contact directly
without contacting other parties.
2.
Accessibility
Easy and fast access to complaint channels.
3.
Responsivenes
Responsiveness of the company to respond to incoming customer complaints properly,
quickly and precisely
4.
Fairness and Objectivity
Then whether we have made steps to follow up the complaint based on the principles of
fairness and honesty, both in terms of results, procedures and interactions.
5.
Customer focus approach
Are all activities, procedures, attitudes and behaviors when handling compail already aimed at
customer satisfaction? for that it needs the involvement of all parties in the company.
6.
Continous improvement
Make every customer complaint a source of improvement. Make it a learning material, learn
from past experiences to be a lesson in the future.
DECISION SUPPORT SYSTEMS IN BUSINESS
9.1.
SPK in Business Decisions
Decision Support System, hereafter shortened to SPK, is generally defined as a system
capable of providing both problem-solving and communication capabilities for semi-
structured problems. Specifically, SPK is defined as a system that supports the work of a
manager or group of managers in solving semi-structured problems by providing information
or suggestions leading to certain decisions (Hermawan, 2005).
Decision-making is the primary function of a manager or administrator. Decision-
making activities include identifying problems, finding alternative solutions to problems,
evaluating these alternatives and choosing the best alternative decision. A manager's ability to
make decisions can be improved if he or she knows the following and master the theory and
techniques of decision making. With the improvement of the manager's ability to make
decisions, it is hoped that the quality of the decisions he makes can be improved, and this will
certainly increase the work efficiency of the manager concerned.
The term structured decision system (SDS) is used to describe systems designed to
help managers solve specific problems. The emphasis is on the word help. A DSS is never
intended to solve a problem without help from the manager. The basic idea is for the manager
and the computer to work together to solve the problem. The type of problems that can be
solved are semi-structured problems. The computer can solve the unstructured part.
Since 1971, DSS has been the most successful type of information system and is now the
most productive problem-solving computer application.
9.2.
Group SPK in Business Pattern
The various committees, project teams and task forces that exist in many companies are
examples of the group approach to problem solving. Recognizing this fact, system developers
have adapted DSS to group problem solving.
Decision Support System (DSS) or SPK (Decision Support System)
•
It is an interactive computer-based system aimed at helping decision makers use data to
identify and solve problems and make decisions. decisions. They are: Heuristics and
mathematical models.
•
A computer-based system that provides both problem-solving information and
communication skills in solving problems Group Decision Support System (GDSS)
•
An interactive, computer-based system that facilitates unstructured problem solving by a
series of decision-makers working together as a group. Primarily a group of managers, in
analyze problem situations and in group decision-making perform tasks.
•
Group Decision Support Systems that seek to improve communication among group
members by providing a supportive environment and supporting decision makers with
GDSS software called groupware.
9.3.
DDS Model
When a DSS is first designed, it generates customized and periodic reports and
outputs from mathematical models. These customized reports contained responses to queries
to the database. Once the DSS was well established, capabilities that allowed problem solvers
to work together in groups were added to the model. The addition of groupware software
allows the system to function as a group decision support system (GDSS). Most recently,
artificial intelligence capabilities have also been added along with the ability to engage in
OLAP.
Group decision support systems (DSS) are interactive, computer-based systems that
assist decision makers in using data and models to solve unstructured problems. It assists
management decision-making by combining data, complex analytical models and tools, and
user-friendly software into one powerful system that can support semi- or unstructured
decision-making.
DSS combines the intellectual resources of an individual with the capabilities of a
computer in order to improve the quality of decision-making.
DSS is defined as an adjunct to decision makers, to expand capabilities, but not to replace
management judgment in decision making.
In a study, Steven S. Alter developed a taxonomy of six types of DSS based on the
level of problem-solving support. The type of DSS that provides slightly higher support
makes it possible for him to analyze the entire contents of the file regarding the budget
absorption rate of other related units. An example is an employee's monthly salary report
prepared from the salary file.
DSS also allows managers to see the possible impacts of various decisions taken
which is called a model that can estimate the impact of a decision.
DSS is intended to complement management information systems in improving
decision making. Management information systems primarily present information on activity
performance to help management monitor and control activities.
The format or form of these reports is generally predetermined (standardized).
Sometimes these management information system reports are exception reports, i.e. they only
highlight special circumstances. Traditional management information systems generally
present hard copy reports. There are two known types of DSS, namely:
•
Model-driven DSS and Data-driven DSS.
•
Model-driven DSS is a stand-alone system separate from the organization's overall
information system. This DSS is often developed directly by individual users and is not
directly controlled from the information systems division. The analytical capabilities of
this DSS are generally developed based on existing models or theories and then
combined with user interfaces that make the model easy to use.
An example of this model-driven DSS used in shipping companies is voyage estimating
decision support systems.
Data-driven DSS, analyzes a large amount of data that exists or is incorporated in the
organization's information system. This DSS helps the decision-making process by enabling
users to obtain useful information from data stored in large databases.
Decision Support Systems include various components that are included in this support
system, namely:
•
DSS database:
A collection of running or historical data from a number of applications used to query and
analyze data. This database can be a PC database or a massive database.
•
DSS software system:
A collection of software used to analyze data, such as: On-Line Analytical Processing
(OLAP) tools, datamining tools. These models can be physical models (workspace design
models, parks, and airplane models), mathematical calculation models (such as: equations,
algorithms, annuities, loan interest installments), or verbal models (such as: a description of a
procedure for writing a work order).
9.4.
The role of GDSS in solving problems
The GDSS is used to identify more problem-solving alternatives and also as a a good means
of communication between group members.
In this sophisticated era, group decision support systems are very useful for an organization or
group in holding meetings and making decisions. Group decision support systems have
several advantages, including:
1. Anonymity. The ability to exchange ideas or preferences anonymously in a group
decision support system environment encourages increased participation by group
members and consequently more information is shared. Participants are no longer afraid
of being laughed at for "stupid" comments, they are also more willing to express
opinions that conflict with other participants or their superiors.
2. Parallel communication. In oral meetings, people have to listen to others speak and
cannot pause to think, group decision support systems allow everyone to "talk" in
parallel (typing and exchanging written comments simultaneously over a computer
network). In an oral meeting Generally, each person only has a few minutes to express
ideas from the entire meeting such as when using a group decision support system.
Parallel communication also contributes to increased group participation and synergy.
Group synergy occurs because other group members will be able to respond to the
proposed idea in different ways, as each participant has a different level of
understanding of the information. In addition, the group as a whole will be better at
identifying errors in an idea than the person who proposed the idea. Reading comments
and providing creative stimulation to others in the group. Criticism is more easily
accepted because it is the idea that is being criticized, not the originator. All of these
factors contribute to increased satisfaction and increased productivity because the group
is more likely to think of ideas as the group's ideas rather than their own individual
ideas because all the ideas have been combined.
3. Automated record keeping. Group decision support systems automatically record
comments, votes and other information shared by a group to disk files. This automatic
discussion log supports the development of organizational memory from meeting to
meeting. In addition, there is no need to take notes manually. Participants in oral
meetings sometimes forget what was said earlier in the meeting and therefore may
forget to comment on the issue under discussion. Finally, in oral meetings, participants
often fail to understand what is being said or may not be able to process information
quickly enough to participate effectively. In meetings using a group decision support
system, participants may spend more time reading recorded comments to better
understand their meaning.
4. More structured. Group decision support systems can provide better structure to
discussions than oral meetings, keeping participants focused on the meeting making it
harder to deviate from the problem-solving cycle and make incomplete or premature
decisions. A group using a group decision support system stays focused on the problem
at hand, and will not discuss or chat about other topics with friends or people next to
them.
5. Due to anonymity, parallel communication, and automatic recording, it leads to new
advantages or benefits. By using a group decision support system, participants in the
group experience greater satisfaction and productivity levels increase, as the group
decision support system shortens meeting time and is able to make better decisions.
FUZZY LOGIC IN MAKING BUSINESS DECISIONS
10.1.
Fuzzy Logic
The word fuzzy is an adjective that means vague or unclear. Fuzziness or vagueness always
encompasses human daily life. (Kusumadewi, 2004). Fuzzy logic is an appropriate way to
map the input space into an output space. (Kusumadewi, 2004). Fuzzy logic is an appropriate
way to map an input space into an output space. For example:
1. The restaurant waiter provides service to the guest, then the guest will give an
appropriate tip for the good or bad service provided;
2. You tell me how cool you want the room to be, I will adjust the rotation of the fans in
the room.
3. The taxi passenger tells the taxi driver how fast the vehicle is going desired, the taxi
driver will adjust the taxi's gas footing.
The reasons for using fuzzy logic are as follows:
1. The concepts of fuzzy logic are easy to understand. The mathematical concepts
underlying fuzzy reasoning are very simple and easy to understand.
2. Fuzzy logic is very flexible.
3. Fuzzy logic has tolerance for imprecise data.
4. Fuzzy logic is able to model highly complex nonlinear functions.
5. Fuzzy logic can build on and apply the experiences of experts directly without having
to go through a training process.
6. Fuzzy logic can cooperate with control techniques
7. Fuzzy logic is based on natural language. Fuzzy logic uses language expressions to
describe variable values. Fuzzy logic works by using the degree of membership of a
value which is then used to determine the value of the variable used to determine the
results to be achieved based on predetermined specifications such as: Fuzzy Variables,
Fuzzy Sets, Talking Universes, and Domains of fuzzy sets.
10.2.
Fuzzy set
A crisp set A is defined by its items. If a € A, then the value associated with A is 1.
However, if a is not a member of A, then the value associated with a is 0. The notation A =
{x|P(x)} indicates that A contains item x with P(x) true. If X is a characteristic function of A
and property P, then it can be said that P(x) is true, if and only if X (x)=1 (Kusumadewi,
2004).
Fuzzy sets are based on the idea of extending the range of a characteristic function
such that it will include real numbers in the interval [0,1]. Its membership value indicates that
an item in the universe of speech is not only worth 0 or 1, but also the value that lies between.
In other words, the truth value of an item is not only true (1) or false (0) but there are still
values that lie between true and false (Kusumadewi, 2004). Fuzzy sets have 2 attributes,
(Kusumadewi, 2004) namely:
1. Linguistics is naming a group that represents a certain state or condition using
natural language. Example: cheap, medium, expensive.
2. Numeric is a value (number) that indicates the size of a variable. Example: 100, 500,
1000, and so on
10.3.
Fuzzy Membership Function:
The membership function is a curve that shows the mapping of input data points into
their membership values (often referred to as membership degrees) which have an interval
between 0 and 1. There are two ways to define the membership of a fuzzy set, namely
numerically and functionally. The numerical definition expresses the membership degree
function as a vector of quantities that depends on the level of discretization. For example, the
number of discrete elements in the universe of speech. Functional definitions state the degree
of membership. analytical expression limits that can be calculated. A certain standard or
measure on the function membership is generally based on the universe X of real numbers:
Linear Representation:
There are two possible linear fuzzy sets: The increase in the set starts at the domain
value that has zero membership degree [0] moving right towards the domain value that has a
higher membership degree. The linear ascending function (right shoulder) is formulated as
shown:
10.4.
Fuzzy Multi Criteria Decision Making Method:
Multiple Criteria Decision Making (MCDM) is one of the most widely used methods
in the area of decision making. The goal of MCDM is to select the best alternative from
several mutually exclusive alternatives on the basis of general performance in various criteria
(or attributes) specified by the decision maker (Chen, 2005: 10). There are two basic
approaches to the MCDM problem, namely Multiple Attribute Decision Making (MADM)
and Multiple Objective Decision Making (MODM) (Kahraman, 2008:1; Tseng and Huang,
2011:1). MADM makes decisions with respect to multiple attributes that sometimes conflict
with each other, whereas in MODM the number of alternatives is infinite and the trade-offs
between criteria are described using continuous functions (Kahraman, 2008:2).
Most MCDM problems in real practice involve not only quantitative but also
qualitative information, which is uncertain. In this case, the MCDM problem should be
considered as a fuzzy MCDM problem that involves objectives, aspects (dimensions),
attributes (or criteria) and possible alternatives (or strategies) (Tseng, 2013) (Huang, 2011:2),
MCDM problems are solved using techniques in the field of artificial intelligence and in
recent decades have become an intensive study of soft computing because they involve fuzzy
set theory.
Fulop (2005) states that, in general, the decision-making process includes decision
steps including:
1. Problem identification
2. Compiling preferences
3. Evaluating alternatives
4. Determine the best alternative.
Based on the description above, some things that need to be considered in the MCDM
problem are: 1). Alternatives; 2). Criteria; 3). Preferences; and 4). Decision-making
tools/techniques. Suppose there are m criteria (C1, ..., Cm) and n alternatives (A1, ..., An).
MCDM problems are usually represented in the form of a decision table as in Table 2.1
(Fulop, 2005).
The value aij shows the performance score of alternative Aj on criterion Ci which is
the preference of the decision maker. Each criterion has a weight wi which indicates the
importance of criterion Ci in the decision-making process.
There are several kinds of preference formats that decision makers can give in
MCDM. Chen (2005:108) mentions five ways to evaluate expert preferences in decision-
making problems: 1). Ordering preference; 2). Fuzzy preference relation; 3). Multiplicative
preference relation; 4). Utility function; and 5). Linguistic variables with conversion functions
to other forms. One of the most widely used preferences in assessment is in linguistic format.
For example, a decision maker gives preference to 4 alternatives {A1, A2, A3, A4}
respectively A1 = "Very Good", A2 = "Good", A3 = "Fair", A4 = "Poor".
Differences in preference formats by individual and group decision makers for criteria
are common in MCDM problems, because each criterion can have different units of
measurement. The different dimensions of the criteria can be solved by the normalization
process, which aims to obtain a scale of values that is compatible with each other
comparability. Various techniques of normalizing preference values against criteria have
become part of MCDM methods (Turskis and Zavadskas, 2010).
In its development, the MCDM method is widely applied in educational assessment.
Many of these applications are related to efforts to conduct assessments that are more reliable
and describe student performance fairly. One of the interesting assessment problems to be
solved using the MCDM method. Affective aspects are usually assessed from observations of
students' daily attitudes and behaviors, therefore, it is possible that the assessment of this
aspect is very subjective and contains uncertainty. Affective assessment usually involves
information that is more linguistic than numerical. In general, there are 5 linguistic variables
used in the assessment, namely SB="very good", B="good", C="sufficient", K="less', and
SK="very less". In MCDM, the affective aspect assessment problem involves linguistic
information as the decision maker's preference, which can be represented in a decision matrix.
