Project Report

Six Sigma- Green Belt Programme SQC & OR Unit Indian Statistical Institute, New Delhi

Gaurav Singh

Project Guide Rina Chakraborty

Executive Summary
This paper consists of a project on Accounting. The project is set in the backdrop of increased demand for better customer (internal & external) service thorough timely release of online system (Micros-Fedelio) after auditing.

The entire job focus was to check revenue losses that might have occurred during the transactions.

The revenue generating departments viz laundry, spa, banquet, bar, restaurants, makes entries of all the transaction in micros. This information is transmitted to Fidelio through an interface. Relating with the job focus the whole concern was to filter all transactional information and pass it on to the Accounts department for proper recording.

The job demanded the use of certain automated auditing process (indexing etc). During this process of automated auditing the major concern was to keep all other users of (Micros-Fedelio) the software logged off and not allowing any further entries or information flow. The system was again provided back to the departments after successful completion of Auditing. As a result front office had no information about the total bills of a guest who is checking out during the auditing process.

This project focuses on the timely delivery of the system to all the concerned departments by avoiding delays due to software failure and preventing intruders.

The endeavor is to strike a balance between timely deliveries of system (VOC) and ensure proper accounting (VOB).

Problem statement
Non-availability of system on time data.

Project scope
As mentioned earlier non availability of system results to revenue loss for the organization. If this problem is addressed properly it will help in revenue increment.

Critical tools
Pareto chart fish bone diagram Failure mode effect analysis control chart

Results
It has been observed that 6.52% is the frequency for failure.

Conclusion & Recommendations

Define Phase

Problem statement
Non-availability of system on time. The ideal time for making the system available is 4 am but it can be tolerated till 4:30am.

Project scope
As mentioned earlier non availability of system results to revenue loss for the organization. If this problem is addressed properly it will help in revenue increment.

Audit and systems

Front Office Guest Room Restaurant Laundry

This is a schematic representation of how the information of various types of consumption by a customer is distributed among various departments. System & audit department is responsible for posting all room charges. The consumptions made by an in house guest in any outlet is posted directly to his room along with his room charges. During check outs front office takes out the actual bill (room charges + other consumptions) through Fedelio, and the bill is presented to the guest for payment.

MICROS

(At revenue generating heads) Interface FIDELIO

(Used by front office, Audit & systems) Interface Oasis (Final accounting software) This shows the flow of information within the softwares. Most of the departments use Micros system while departments like Front Office, Systems and audit uses Fidelio, oasis is only used by Accounts for final documentation. The flow of information between every software is intermitted by an interface.

The current system pays attention to the most frequented VOC (voice of customer) System is not available (due to delay in audit) while a guest is on the verge of leaving the hotel. On the other hand VOB (voice of business) is responsible for loosing revenue due to non availability of system. Both VOC & VOB talks about non availability of system- constitute the take of point for this project.

Key measures
Impact on key business

Following threat-opportunity matrix shows some of the short and long term business impacts.

Harassment of customers Revenue loss

Increased customer satisfaction Increase in revenue

Business value defeated Loss of good will

Creation of business value Creation of good will

Short Term

Long Term

Threat

Opportunity

The Measure Phase

Performance
Identification of kpivs & kpovs

Quantify current performance level: Issues Number of days System released on time System release 3 1 11 2 Jan 31 28 Feb 28 27 Mar 31 20 April 30 29

delayed DPMO
96774 35714 354838 64516

The month wise data and the current run chart shows that the defect level in the month of March is abnormally high. This was mainly due to some exceptional machinery failure. Except for the data of March the rest of three month data is considered in arriving at the process defect.

Current defect level

Total number of System release delays ( All three months) Total number of days in 3 months (excluding the month of March)

*106

Total number of System release delays (computing all three months) = 6 days Total number of days in 3 months (excluding the month of March) = 92 days = 65217 this is approximately equal to 3.009 Sigma

Targeted DPMO = 50 % of the current DPMO

Targeted DPMO = 32608 this is approximately equal to = 3.345 Sigma

Current DPMO Targeted DPMO

65217 32608

Analysis and Findings

The project team uses a fishbone diagram to identify all the input variables that could be causing the problem of interest. Once we list out the input variables we can identify the critical few process input variables to measure and further investigate.

Non delivery of system on time

MACHINE

Training Experience intrusions

Number of os NO of systems

ENVIRONMENT Seasons No of employees Experience of employees. Lack of security in the software to stop log in

MATERIAL Rush at the restaurants Stability of accounting system

METHOD

Closing of RGHs Pre night audit corrections

MEASUREMENT Revenue loss Customer dissatisfaction

Starting of CRIS delayed

After listing of all the input variables, some of the key process input variables (KPIV), Lack of security in the software to stop log in, pre night audit corrections are identified as the main culprits. The entire focus is turned to these two problems referred to as vital Xs. dentifying vital Xs Potential Xs

Jan

Feb

April FREQUENCY

1)

Lack of security in the software to stop log in.

2)

Pre night audit corrections

3)

Rush at the restaurants

TOTAL

It can be observed that out of a total of 6 occurrences 4 are attributed to lack of security in the software to stop log in.

Pareto chart of potential Xs

It can be seen that the problems Lack of security in the software to stop log in and pre- night audit corrections together form more than 80% of the assignable cause. Hence we are targeting these two most prominent problems, solutions of which will help reduction in the problem and vis--vis increase the sigma level of the operation.

High Within control Lack of security in the software to stop log in.

Medium

Low Pre night audit corrections

Beyond control

Rush at the restaurants

Analyzing vital Xs

The focus now converges into two controllable factors affecting replenishment performance

Lack of security in the software to stop log in Pre night audit corrections

Lack of security in the software to stop log in

It has been observed that the major reason for the delay in releasing the system is log in after the audit starts off. This results in corruption in the system as a result, finally the entire process of night audit is delayed and subsequently the system delivery does not take place on time. Out of the total number of observations 66.67% pertains to this problem.

It is also observed that during the month of January, out of a total 3 failures 2 are assigned to this current problem of Lack of Security in the software to stop log in.

As a result any person entitled to use this software can log into the system even after the rights for the access has been withdrawn and the night audit is hampered.

Pre night audit corrections.

The time of release of system to the users on time is also commanded by the number of accounting errors that the auditor has to handle before moving on to the night audit procedure. The problems may vary from incorrect room rent charged to wrong settlement type (credit card in place of case or vice versa). In all the cases of error (since this is not under the current scope of the defied problem) it can be assumed that they happen because of lack of proper training and development and excess work load.

Improvement
Failure mode and effects analysis

This process is used to understand and quantify the relative importance of the major concerns and its development components

Concerns

Ratings

Improvement of software

Training

Increase of manpower.

Lack of security in the software to stop log in.

5 (25)

0 (5)

0 (0)

Pre night audit corrections

2 (8)

2 (8)

3 (12)

Rush at the restaurants

0 (0)

2 (2)

4 (4)

Lack of Training

1 (2)

4 (8)

1 (2)

Improvement of software

Improvement of software is one of the major issue of improvement. The software lacks the facility of restricting the user to access datas or to make entries while auditing is being conducted. The software also lacks the facility of at least warning the user before allowing access incase the user is trying to access it during the time audit is being conducted.

Therefore the software has to include a warning signal to user when they try to log in during night audit, and also restrict the user if necessary.

There should be a window dedicatedly for information purpose and it can be accessed even during audit. This system will help checking revenue loss by providing updated billing information to the front office and subsequently to the customers.

Pre night audit corrections

The timely delivery of system also depends on how many errors has crept in during the day, the auditor first has to resolve these problems before he moves on to automated night audit procedure.

The errors are mostly due to lack of proper training and excess work load. The training related issue can be taken care of by training the staff in using the software by conducting mock transaction. As far as the workload is concerned it can be handled through proper recruitment process and decreasing the per day working hours.

Previous sigma level Current sigma level

3.009 3.249

Time for system access restored for the month of May

DAY 1 2 3 4 5 6 7

Time (am) 4 4:11 4 4:02 4:05 4 4

8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

4 4:07 4:16 4:01 4 4:12 4:13 4:04 4 4:17 4:22 4:45 4:08 4:16 4:03 4 4 4:07 4 4:11 4:23 4:14 4 4:27

Calculation of improved Sigma level:

Issues Number of days

May 31

June 30

July 31

August 31

System released on time System release delayed DPMO

30

29

30

29

32258

32258

32258

Total number of System release delays ( All three months) Total number of days in 4 months

*106

Total number of System release delays = 4 days Total number of days in 4 months (excluding the month of March) = 123 days = 32520 this is approximately equal to 3.24 Sigma.

Control phase
For the purpose of verification of control a simplified control chart (attached in the graph) is implemented with a tolerance limit of 4:30 am, the ideal time for delivery of system remains at 4 am. in the control chart there are three upper limits defined (lower limit does not exist) as 4am ideal time, 4:15am minimum tolerance time, 4:30 maximum tolerance time. For a one month data (may) if plotted on a graph it will have 24 points within the minimum tolerance limit (4:15am), 6 points within the maximum tolerance time(bold points) and 1 point above the maximum tolerance time (4:30 am, point and a circle)

Every time the system is delivered above the minimum tolerance limit the observer can become alert and adjust some KPIVs to regain desired performance. If the graph shows a point above the maximum tolerance limit it means the system requires adjustments it might have fallen out of control.

Conclusions
Restatement of quantified project results: Operating DPMO: 65217 Operating sigma level: 3.009 No of days under observation: 92 days

Projected results: Targeted reduction of DPMO: 50% Predicted DPMO: 32608 Predicted sigma level: 3.34 No of days under observation; 123 days Key performances input variables Lack of security in the software to stop log in. Pre night audit corrections.

Recommendations
Lack of security in the software to stop log in: redesign the software to secure it from intrusions during audit procedure. Pre night audit corrections: to train staff properly in order to reduce mistakes.

Appendix
Six sigma green belt training module(ISI) All about Six Sigma- Warren Brussee