Loading presentation...

Present Remotely

Send the link below via email or IM

Copy

Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

DeleteCancel

Make your likes visible on Facebook?

Connect your Facebook account to Prezi and let your likes appear on your timeline.
You can change this under Settings & Account at any time.

No, thanks

Cashier Policy Optimizer for Large Retail Store

No description
by

areej zaki

on 9 September 2013

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of Cashier Policy Optimizer for Large Retail Store

Cashier Policy Optimizer for Large Retail Store
Team Members
Areej Mohamed Zaki
Nada Mohamed Mohsen
Nourhan Alaa-Eldin
Yosra Abdel Karim
Presented by
Areej Mohamed Zaki
Sponsored by
Agenda
1. Introduction
2. Simulation
What is simulation?
Why simulation?
Advantages vs. disadvantages
Types of simulation
Simulation Application Areas
3. Queuing Systems
Agenda
4. Simulation Model Implementation
1.Understand the system
2.Problem formulation
3.Input data collection and analysis
4.Formulating the model
5.Model translation
6.Verification
7.Validation
8.Design experiments
9.Output analysis
10.Recommended results and documentation
Introduction
Management science technique for analysis and study of complex systems.
What is simulation?
The process of designing a logical model of a system and then conducting computer-based experiments with the model to describe, explain, and predict the behavior of the system.

Why Simulation?

Better understand the expected performance of the system.
Simulation
Advantages
Allow to perform un-
limited num. of exp.
Easier to understand
High Speed
Tolerates for using
un-deterministic variables.
Generally less costly
Disadvantages
May not provide you with the optimal solution.
Time to construct model will be longer.

Types of simulation
A
static
simulation model is a representation of a system at a particular point in time ; as a Monte Carlo simulation.

Types of simulation

A
discrete
event is a state of system that changes only at discrete points in time ; as Queuing System.

Simulation Application Areas
Robot simulators for the design of robots and robot control algorithms.
Queuing systems
Data requirements (Inputs)
Inter-arrival time distribution.
Service time distribution.
Number of servers.
Queue discipline.
System capacity.
System Performance measures
Expected number of customers in system.
Expected number of customers in queue.
Expected time in system.
Expected time in queue.
Server utilization.
Probability of numbers of customers in the system.
Simulation Model Implementation
In order to implement a simulation model we have to follow the following 10 steps.

1. Understand the system.
2. Problem formulation .
3. Input data collection and analysis.
4. Formulating the model.
5. Model translation.
6. Verification.
7. Validation.
8. Design experiments.
9. Output analysis.
10. Recommended results and documentation.
1.Understand the system

To understand the system we had several meetings with hyper one representative to know
How does the system work?
What are the main problems in the system?

Cashier simulation.
Forecasting of sales especially seasonal ones.
Monitoring the flow using RFID (Radio Frequency Identifier).
VMI (Vendor Managed Inventory).


2.Problem formulation

Problem statement
Like any retail store, the Egyptian wholesaler/retailer HyperOne faces a serious problem with its cashiers (Point of Sale or POS) which can be easily recognized with the long queues at the POSs affecting
Problem causes and effects
In our problem there are several causes that lead to the effect of poor cashier performance which reduce the customer satisfaction and blocks the shopping area affecting the flow of customers in Hyper One. The following causes are illustrated in the following Fishbone diagram.
System definition
Our system is classified as Discrete, Stochastic (input random) and open (interact with others) system. Also our system is said to be a terminating system.

System description

Physical View
Hyper one consists of 2 floors each floor approximately consists of 27 cashiers. Our narrow system of interest includes only the first floor which has 25 normal cashiers and 2 less than 10 items cashiers.
Swimlane diagram features divisions or "lanes." Each lane is assigned an actor (which may be an individual, department, division, group, machine, entity, and so on), or even a phase or stage in a process, that is responsible for the activity or work described in the lane.
System description

Solution Statement
Develop a Cashier Simulation Model:
Identify the key of performance.
Construct a simulation model that represents customer flows.
Assess staffing policies at point of sales.
Select a policy balancing staffing costs against benefits.
System Framework

3.Input data collection and analysis
After several visits to HyperOne we determined the input data which are derived from the system
Input data collection
and analysis

The output data which are concluded from the system

Average time in the system.
Average time in a queue (Average waiting time).
Average utilization rates.
After knowing what are the basic input and output data in the model , we conducted the sheet that will include all the data collection elements.
Data Division

As we had access to the Database of hyper one. The data was divided into
Data driven from the database.
Data to be collected manually.

Data driven from the database

Data to be collected manually

Consisted of two separate sheets
Data Cleansing

After collecting all the needed data we had to make data cleaning
To consider only the required data.
To join between the data from the database and the data that was collected manually.


SAS Enterprise Guide
The last step in the data collection part was to analyze the data for introducing it to SAS Simulation Studio, to do so we used SAS Enterprise Guide so as to know the type of each data element and the measures that will be needed.

This Figure is the Scanning time for less than 10 cashiers whose experience is high.

4.Formulating the model

We obtained a fundamental understanding of the system logic which we did by constructing a high level flow chart that also illustrates how the input and output data play a role in the model.

5.Model Translation
Simple model

Less than 10 cashiers model

Detailed model

Packing time

Total Service Time

Error Time

Payment Time

Error example
6.Verification

To verify that the model was going in the right track we conducted the simple model


7.Validation

Validation was defined as the process of ensuring that a model represents reality at a given confidence level.
However, no matter how good and enhanced your model is, still the model may not be suitable for conducting analysis. This result from:

Assumptions


We make assumptions due to lack of knowledge certain assumptions may have to be made with respect to the system components, interactions, and input data.

The behavior of the customer
We assumed that the customer will always choose the shortest path.

Cashier experience
We assumed that the cashier may have either excellent or medium experience based on the scanning time he takes in scanning each item and his experience in the feild.

In case no bagger
We assumed that when a bagger doesn’t exist the cashier stops the queue and block any other customer from entering the service until he helps the current customer to pack his goods.

Simplifications
We had to make deliberate simplifications in the model of the system in order to finish a simulation project in the allotted time.

These simplifications where:
Customer renege.
Bagger as a decision variable.

Limitations
There are likely to be many limitations with respect to being able to model complex system.


Oversights
If there is any complexity to the system being modeled, it is extremely probable that the practitioner is going to inadvertently overlook one or more critical system components.

System Errors:
Errors in the POS machine itself such as:
The scanner maybe very slow.
The POS machine itself may be operating very slowly.
The touch screen respond time is slow.

Oversights
8.Design of Experiments

This feature in SAS Simulation Studio provides a powerful facility to automate running a wide range of simulation scenarios and the capability to conduct full design of experiment testing that provide us with the information necessary, decision recommendations with respect to the project objectives.

Less than 10 cashier DOE

Responses
Factors


Responses of the detailed model
(only 20 points)


9.Output Analysis

After constructing and running the model, analyzing the data and observing the system as a whole we came to a lot of conclusions from the Design of Experiments that helped us to manipulate in the factors and check their responses in order to draw our final conclusion that was based on our study, data analysis output.

SAS Simulation Output

Two graphs were made on SAS Simulation Studio software
One showing the average waiting time of the customer in all the queues.
The other showing the average utilization of the queues which either the queue is in action/working or not.


10.Recommended results and documentation.
Conclusion
The project, further, showed how simulation can be linked to available systems and databases allowing online analysis of the situation. To complement the presented model two more components are being developed:


This project shows the different tools of operation research, simulation, data , design of experiments and modeling.
Developed model and framework can be productized and used in other retailers.
Base for enhancing tool for decision making.
Up to our knowledge scale with SAS in Egypt .

Potential
Demo
Demo
A technique that imitates the operations of a system as it evolves over time.


Simulation refers to a broad collection of methods and applications to mimic the behavior of a system.

Improve the operations of the system.

Understand how the system works.
Design certain operations if it don’t exist.

Provide managers or controllers of the system with readily available aid for day-to-day operations.
A
dynamic
simulation is a representation of a system as it evolves over time.

A
Continuous
event is a state of system that changes continuously over time ; as Level of fluid in tank.

All inputs must be in distributions
What is the chosen problem to be handled?

The problem that we have chosen was cashier simulation problem.
Cashier simulation is the main problem Hyper One is currently facing, their objective is to minimize the service time and the queue length, and also they need to know the maximum number of cashiers and shifts for everyday.
Service time.
Reducing customers’ satisfaction.
Blocking the shopping area.
Affecting the flow of customers in hyper.
The queuing system is characterized by three components Arrival Process, Queuing Discipline and Service Mechanism which we understood very well.
Inter-arrival time.
Service time.
Error type and time.
Failure rates.
Payment time.
Packing time.
While the other sheet is for the action that takes place on the POS itself and it includes all the other elements.
one will be for calculating the inter arrival time and this is calculated by monitoring the number and time of customers entering the queue.
Only shows the main logic of the system.
Consisted of 1 queue.
Without any details.
Illustrates the flow of the cashier in the real system.


Then we tested on it the data that was collected and started to expand this simple model into the full detailed one.

To simplify the complex or thought to be insignificant internal detailed workings of the process.

One floor only.
HyperOne cashiers.
All occasions.

Inter Arrival time
we didn’t collect the inter-arrival time for all the cashiers in one shift to represent the inter-arrival time in reality, we made the collection in different shifts and gathered them together.
 
Conveyer error:
We also found that some POS has errors in their conveyers that make the customers have to push their goods manually to the scanner.
Sometimes these conveyers break down during the service time of a customer and the cashier tries to fix it and this takes up to 10 minutes.
(1) A cost module to provide cost-benefit analysis.
(2) A user interface to ease the interaction between the management and the tool.
Traffic engineering to plan or redesign parts of the street network from single junctions over cities to a national highway network, for transportation system planning, design and operations.
Full transcript