|
Process Improvement Project |
|
Our process improvement project began with River Club Apartments where a member of Bulldog Consulting resides. The apartment complex is relatively big, containing over 200 units, so a variety of potential improvement areas were identified. After presenting the project to the complex manager, these improvement areas were narrowed and it was determined that the gate entry/exit system was the process most in need of improvement. River Club contains a card access system allowing entry into the complex. A Guest scrolls through a list of residents, which then calls the tenant. Access is granted to the guest upon consent from the tenant. Residents enter through the gate simply by using their access card. One exits through the other side of the gate system by pulling their car up close to the gate. This gated environment is extremely important to a complex comprised of mainly college tenants because of its use as a recruiting element with parents. The problem presented to Bulldog Consulting was to reduce the number of operating malfunctions with the gate entry/exit process and to lower the amount of time the gate is inoperable. |
|
After establishing our goal to minimize operating problems with the gating process, our group began devising a flow chart to determine how the gate entry/exit process works. Kate Jones, the manager at River Club Apartments, thought the main reason for the breakdowns of the gate was due to the slow response time of the security service that ran the gate entry/exit process. If the security company doesn't repair the gate as soon as it breaks down, more serious problems arise from tenants and guests forcing their way into and out of the apartment complex. After devising the flow chart, Bulldog Consulting suggested to River Club Apartments that they get a local security service, since the current is out of Atlanta, to run the gate entry/exit process. With a local security company, response time will be greatly improved, same day service might even be possible. Even though Bulldog Consulting felt that the suggestions would help, we didn't feel that an Atlanta based security company was the root cause; therefore, Bulldog Consulting started a checksheet of what we thought could be the root cause. Through discussions with Kate Jones and other workers at the complex, it was discovered that most gate malfunctions were due to hydraulic failure, which occurs from drivers running into the gates. Kate also noticed a number of people trying to sneak through the exit gate as it closes. So one evening from 4:00-8:00 we observed the gates checking two problems: people entering through the exit gate, and people hitting the entry and exit gates with there bumpers. The result of the checksheet was that people entered through the exit five times more than the gate was hit. From this data we determined the possible root cause to the gate entry/exit problem might be people entering through the exit; however, we were not quite sure if that was it since we know the gate doesn't malfunction unless it is hit. Thus we made an Ishikawa diagram to determine the exact root cause. The Ishikawa diagram showed some potential problems in the gate entry/exit process such as inoperative access cards, and problems with the card reader; however, the Ishikawa diagram in fact showed, on the largest branch of possible causes, people entering through the exit gate to be a root cause. This is the root cause because sensors to open the exit part of the gate are only located inside the gate, so when people attempt to enter through the exit gate, they have less time to get to the sensors before the gate closes. Thus the gate gets hit from time to time causing hydraulic failure. |
|
At that point in the project, we began brainstorming for improvements. Ideas included putting up signs, putting speed bumps outside the exit gate, and better illuminating the exit gate. The idea that was finally agreed upon though was putting up a barrier gate that would only allow one car to enter through at a time. We knew this idea would be more expensive than some of the other ideas, but after discussing it with the complex manager, it was concluded it would be the most efficient. The complex manager contacted the security company, and it was determined that this idea would be feasible. |
|
Our group decided that the best way to measure the extent of this problem would be to watch videotapes from the complex security cameras. After discussing with the complex manager we decided that the busiest times were from 12:00 to 4:00 p.m. on Sunday and weekdays and from 10:00 p.m. to 2:00 a.m. on Friday and Saturday. We divided up tapes and measured the number of times people entered through the exit gate, the number of times the gate was hit, and the number times of the gate broke from being hit. Our pre-data results showed that the exit gate was entered 79 times during the week through the exit, twice the gate was hit, and once the gate required maintenance from being hit. The mean, upper control limit (3 st. dev.), and lower control limit (3 st. dev.) were determined for the number of times people entered through the exit gate to achieve a 95% confidence interval for the data. The results from our data collection were plotted on a pre-data control chart. Our method to prove that our implementation is successful is to plot the post-data results on a control chart using the same mean, upper control limit, and lower control limit. An improvement will be seen if the results show the pre-data 95% confidence interval invalid meaning some plots are under the lower control limit. |
|
Because of the financial implications involved with this project, most of the discussions with the security company putting up the barrier gate, were handled by River Club. In coordination with management, we established April 29th through May 1st as the dates that the barrier gates would be put up. We then collected post-data the same way as we did pre-data, observing the videotapes from the same hours. As expected, our data showed vast improvements from the original study. The number of cars entering through the exit gate decreased from 79 cars in a week to 8. Both of the other factors measured were totally eliminated by implementing the barrier gate. The results plotted on the control chart also verify that the barrier gate improved the gate entry/exit process. Five out of the seven days tested resulted in the number of cars entering through the exit gate to be under the lower control limit. The implementation was successful in lowering the times cars enter through the exit gate that is directly correlated to the breakdowns in the gate entry/exit process. |
|
Even though implementation of the barrier gate did not completely prevent people from entering through the exit gate, vast reductions were shown and the gate was not hit; however, with cars still entering through the exit, hitting the gate is inevitable. The management at River Club Apartments was extremely happy with the results and felt that the benefits of implementing the barrier gate would far outweigh the extra cost of putting them up. Overall, we felt our project was a success, and the improvement could be measured in the decrease in the amount of people entering through the exit gate as seen on the control charts.
|
| Executive Summary |
Our group's project involved improving the gate entry/exit process at River Club Apartments. Initially, it was thought that the problems were due to poor maintenance by the security company, but after making an Ishikawa diagram we determined that the root cause was people entering through the exit gate. We concluded that this was the root cause because the most common reason that the gate was broken was due to cars hitting the gate while trying to enter through the exit gate. After reviewing videotapes that monitor the gates we determined 79 cars per week enter through the exit gate. In order to solve this problem, we suggested and implemented a barrier gate, and ultimately, we reduced the number of cars entering through the exit gate to 8 per week.
|
Since the flowchart didn't produce the root cause, further analysis was done.
|
- First a checksheet was performed to help narrow down the possibilities. Enters through the exit gate were included as suggested by Kate Jones, manager River Club Apartments.
|
| Checksheet |
| 4:00-6:00p.m. |
| Summary |
| # that enter through exit gate | x | x | x | x |
| # of times the gate is hit by cars | x | | | |
|
- The checksheet showed that the enters through the exit might be a possible root cause, but we were not sure why.
|
- An Ishikawa Diagram was devised with the gate entry/exit process on the right to determine the exact root cause.
|
- Bulldog Consulting then asked why their are problems with the process and came up with the two branches.
|
- The suggestions resulting from the fishbone diagram would help out with the card entry system problems, so we focused our attention on the other branch.
|
- The Ishikawa Diagram showed that the root cause of this branch to be the number of cars entering through the exit gate.
|
- Bulldog Consulting had several ideas of ways to improve the process, but a barrier gate that would only allow one car to enter/exit at a time was perceived as the best alternative.
|
- As the plans for improvement were suggested to management, we began to collect pre-data from the complexs security cameras.
|
- From the videotapes, Bulldog Consulting was able to measure during peak hours the number of cars that entered through the exit, number of times the gate was hit, and the number of times the gate broke.
|
| Pre-Data |
| Day | 1 | 2 | 3 | 4 | 5 | 6 | 7 |
| # of Enters Through Exit | 10 | 6 | 8 | 11 | 16 | 19 | 9 |
| # of Times Gate Hit | 0 | 0 | 0 | 0 | 0 | 1 | 1 |
| # of Times Gate Broken | 0 | 0 | 0 | 0 | 0 | 0 | 1 |
|
- From the pre-data, a control chart was devised to give the 95% confidence interval for the enters through the exit gate.
|
| Values Calculated |
| UCL | 21.364 |
| Mean | 11.29 |
| LCL | 1.21 |
|
After the barrier gates were implemented, Bulldog Consulting begin the post-data collection.
|
- The results were as follows:
|
| Post-Data |
| Day | 1 | 2 | 3 | 4 | 5 | 6 | 7 |
| # of Enters Through Exit | 0 | 1 | 1 | 0 | 2 | 4 | 0 |
| # of Times Gate Hit | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| # of Times Gate Broken | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
|
- The number of times the gate was hit and/or broken decreased to zero observations for every day. The time cars entered through the exit also drastically decreased.
|
- The data collected from the enters through the exit gate were then plotted on a control chart with the same UCL, LCL, and mean as setup with the pre-data observations.
|
- From the control chart it can be observed that five out of the seven days were lower than the LCL as a result of the barrier gate. Thus the data supports that the 95% confidence interval set up earlier is no longer valid and that a new one would be lower than the current.
|
- The barrier gates implemented to improve the gate entry/exit process did in fact improve the process; however, the barrier gates did not turn out to be able to poka-yoke the error.
|
