INTRODUCTION The aim of this report is to examine the existing business model that is in operation in S;J garage, an understanding of what the inputs and outputs that are required for the smooth running of the business and a weekly activity that mimics the real life situation of the company will be presented. The software that will be used to carry this out is called ARENA; it is a software that has been developed based on the principle of flow-oriented simulation.

It allows for the modeling of real life scenarios and modifications can be made in order to test and an opportunity is given to experiment with what we normally cannot do with the original business because of the cost implication. This is a very great way of increasing overall efficiency of any business. A table will be drawn to illustrate the entities, state variables and events. A flowchart diagram would also be used to show a representation of the overall process flow.

No more academic routine!
Order your coursework right now!


order now

An activity cycle diagram will also be created, this will identify the key processes and the resources that are assigned to them; it will also show what the sequence of activities is like during the process and how the resources’ states change (“idle” to “busy” and vice versa) when an event occurs. What these diagrams will do is to further assist in understanding the situation at S&J. When the model is complete, what the current resource utilization, bottlenecks, costs incurred and revenue received are is analyzed according to what the real life situation is.

Afterwards, amendments will be made to some parts of the model to ascertain if an improvement in its efficiency can be changed. Some of the improvements to be made include: Minimizing the overall service time Achieving a minimum queue at each work Centre. Maximizing the resource(s) utilizations. Minimizing average car waiting time in the overall process. If all of the above can be achieved it will lead to reduction of costs which are associated with the business operations.

Finally, future directions and recommendations will be suggested as regards what can be done to what they have at the moment to increase overall efficiency of the company and ultimately its profitability. 1. 1 AIM The aim of this simulation exercise is to minimize the overall simulation time, achieve a minimum queue at each work center, maximize the resource(s) utilizations and minimize average waiting time. 1. 2 OBJECTIVES The objectives of this simulation include: * To use ARENA software to model the real life business scenario of S&J garage. To create an understanding with the aid of a table, an activity cycle diagram and a flow chart diagram of the core entities and processes carried out with in the business model. * To analyze the utilization of resources and the bottlenecks present in the current system. * To create scenarios, which modify and increase the productivity of the current business model. * To choose the best scenario that gives the best efficiency of the business model. * To suggest further directions, recommendations and suggestions that could give the garage a better competitive edge over their rivals. . 3 ASSUMPTIONS To effectively model this, the following assumptions will be applied:

1) An 8:00am-5:00pm work day will be assumed i. e. a 9 hr work day.

2) In each repair bay there will be 3 workshops to service each vehicle type.

3) The vehicles are counted for record purposes as they arrive at the garage.

4) It is possible to employ more mechanics to work at the repair bays.

2. 0 ENTITIES, STATE VARIABLES AND EVENTS

2. 1 PROCESS/TIME TAKEN Booking In (6 Minutes) Paying (6 Minutes) Repair (Different Times) Wash (10 Minutes)

2. RESOURCE/ NUMBER

Receptionist (1) Admin Employee (1) Mechanics (3) Auto-Machine (1) 2. 3 RANDOMNESS AT BAY 3 3| 1| 3| 5| 4| 1| 6| 2| 6| 4| 6| 3| 3| 5| 4| 3| 5| 6| 1| 1| 6| 6| 3| 1| 4| 2| 4| 4| 1| 5| 2| 6| 3| 5| 4| 6| 5| 3| 3| 5| 1| 3| 1| 1| 2| 6| 1| 4| 2| 3| 5| 4| 4| 3| 6| 6| 5| 5| 6| 4| 4| 4| 1| 5| Table 5: Lorry Repair Process Times at Bay 3: Repair Centre 1 The above data was given in the question and was analyzed thus: First it was copied into notepad and then analyzed using the input analyzer section of the Arena software.

Results are shown below. RESULTS FROM INPUT ANALYSER 3. 0 DATA COLLECTION METHODS All the information that will be necessary in the modeling of the S&J Garage will be gathered using the quantitative data collection method. Quantitative Research is one, which is concerned with numerical quantities and measurements. It may involve a calculation of probabilities and estimates. 3. 1 MODEL VERIFICATION AND VALIDATION

Verification will be carried out initially by looking at the animation of the entities and using the function called the track function to track the entities throughout the system; this method would make it possible to make sure that the entities are actually going to the correct locations according to the flow diagram that has been created. 4. 0 AS-IS SITUATION MODELLING The “As Is” situation arena model in the figure above presents an imitation of the real life production flow process of S&J Garage. The first section of the simulation model is called ‘ARRIVALS INTO GARAGE’ and it models the arrival of vehicles into the garage.

The create module is used to model this and three different create modules are used because there are 3 types of vehicles expected to come into the garage. The entry was modeled using the constant expression to express the inter-arrival times of the vehicles. The assign module was then used to record the time of arrival of the entities into the system. The record module is then used to keep a record of the number of entities that arrive at the system. Thereafter the decide module is used to allocate the percentage of cars that belong to each type of vehicle.

At this point the 3 different vehicle types are split. The assign module was then used again to attach pictures to the entities for easy identification when moving around the system. The next section of the simulation is called ‘BOOKING AND FAULT IDENTIFICATION’ and it models the booking and fault diagnosis of the vehicles after they arrive at the garage. The process module is used to model this and a resource called ‘receptionist’ is placed there to book in the vehicles and attach a fault diagnosis sheet to them. The time taken to perform this duty is constant at ‘6mins’.

The decide module is then used to split the faults into 4 categories as the vehicles are repaired according to these faults. The next section is the ‘PAY POINT’ and here another process module is introduced here to model it. The ‘admin officer’ resource is placed here to assist customers in paying for the services to be rendered and also to guide them to the appropriate repair bay. The decide module is then used again to split the vehicles into their various faults so as to aid them in going to the right bay for repair. In order to decide the ‘n-way by chance is used’ because the faults occur with a uniform distribution i. . there is a 25% probability of occurrence of each fault. The reason for this split is because a different crew of mechanics is responsible for individual faults although sometimes they overlap. Another set of decide modules are used here to split each fault into the three car types again because the car types take different times to complete. The next section is the ‘REPAIR BAYS 1, 2, 3;4’ and these are used to model the repair process that the cars have to pass through. 3 process modules are used in each bay to represent all 3 car types which come into each bay for repair.

Here decision is based on ‘n-way by condition’ which specifies the entity types which are allowed to pass through the true exit and which entity passes through the false exit. The next section is the ‘AUTOMATIC CAR WASH’ and it is used to model what happens to the car after it has been repaired. Usually the cars are dirty after they have been repaired and need to be washed. However, there is a limit to the patience a customer can have to wait for their car to be washed. The decide module is then used again here to show how many cars wait to be washed.

The ‘2-way by condition’ is used here to specify the exit routes of vehicles for the true exit the expression condition is used and the expression {NQ(WASH VEHICLE. Queue). LE. 5} is used and this is to specify to the software that as long as the queue at the car wash is less than 5, more cars are welcome to wait to be washed but as soon as the queue hits 5 or more the vehicles should exit the garage. The last part of the model is the ‘LEAVE GARAGE’ and there is only one module here and it is the dispose module which is used to exit entities from the system.

For cars that have been washed and those that met a queue of 5 cars and could not wait to be washed, their next port of call is the dispose module. This ends the simulation. 4. 3 RESULT ANALYSIS OF THE AS-IS MODEL Discrete Event Simulation Modeling is used to analyze and detect the reasons behind the aforementioned bottlenecks to improve the ability of the supply chain to run in a more efficient workflow without any disturbance. (Al Bazi, 2011) An identification of the bottle necks in the same is needed in order to effectively analyze the as-is model that has been created.

The method which will be used is to outline the key performance indicators of the garage system. These will also serve as a measure of performance of the model. They are: 1. The overall service time. 2. Queue at each work Centre. 3. Resource(s) utilizations. 4. Average car waiting time in the overall process. 4. 4 KPI’S Each of the key performance indicators mentioned above will be treated in depth in this section. Each one will be elaborated upon and the numerical output presented by the simulation software as a result of the simulation run will be presented to further illustrate the analysis. Overall service time

This refers to the total time it has taken for the simulation to come to a complete end. The simulation that has been modeled has 5 replications. The total time for each individual replication will be added together to come up with the overall service time. (Figures are in hours) Replication 1 Simulation Time = 25. 97 Replication 2 Simulation Time = 20. 15 Replication 3 Simulation Time = 28. 01 Replication 4 Simulation Time = 26. 55 Replication 5 Simulation Time = 26. 78 Total Run time: 127. 46hrs Queue at each work centre This refers to the number of entities that are waiting to be served by the resources at each workstation.

It is an important KPI because it helps us to know which resources are the busiest and the work stations that have the most entities to service. Figure: Waiting time of entities at each work centre for the as-is scenario From the waiting time of entities at each work station table and graph above it can be easily seen that the process with the highest waiting time is the Car repair at Bay 2 (Repair CAR 2), a close second is the Lorry repair at Bay 2 (Repair LORRY 2), followed by Car repair at Bay 4 (Repair Car 4), Car repair at Bay 3 (Repair Car 3) and HGV repair at Bay 1 (Repair HGV 1) respectively.

It is on the basis of these that the experiments will be made seeking to reduce them in order to minimize the queue at each work centre. Figure: Number of entities waiting in the queue at each work centre for the as-is scenario From the figure and graph above it is seen that the admin officer has the longest queue followed by the receptionist, the mechanics at the HGV repair at Bay 4 and then the mechanics at the Car Repair at Bay 4 respectively. We will seek to reduce these queues in order to effectively improve the performance of the system. Resource Utilizations

This refers to the usability of the resources and how many of the resources are actually being utilized at each work station the importance of this KPI is to know which resouces are too busy and which are idle. Idle resources will be assigned more jobs while the busy ones will be relieved of some of their duties. Figure: Scheduled utilization of each resouce at the work station From the figure above, it can be detected that the highest utilised resources are mechanics 3, 2 then 1 and what the experiments will do will be to reduce the work done by those resources and assign more jobs to other resources.

This will effectively increase the performance of the system. one thing that is interesting to note however is the fact that even though the other 3 resources seem to be under utilized from the figure below, we can see that the actually handle the most resources. Figure: Total number of entities that each resource works on From the figure above, the three resources that deal with the most entities are the receptionist, the carwash machine and the admin officer. They are all dealing with all the entities that enter into the system.

Mechanic 2 treats the least number of entities. 4. 5 THE EXPERIMENTS (WHAT-IF? ) SCENARIOS These are the experiments to be carried out on the model which seek to improve the overall performance of the garage system by making he key performance indicators better. The scenarios to be used will be such that each scenario will satisfy all the KPI’s. What-If? Scenario 1 From the analysis above it can be seen that Bay 2 has the busiest resources and we can also see that the resource 3 is also very utilised and resource 2 is the least utilised resource.

What will be experimented in this scenario is to reduce the work load at Bay 2 by completely removing mechanics 2 & 3 from that bay to allow them perform better at other bays and adding 2 extra workers (Mechanic 4 and Mechanic 5) for each process in Bay2 that has the queue there to help improve performance at that bay. Once again each KPI will be recalculated in order to see what the changes to the model will do for the overall performance of the model. Overall service time Replication 1 Simulation Time = 16. 97 Replication 2 Simulation Time = 18. 49

Replication 3 Simulation Time = 22. 01 Replication 4 Simulation Time = 20. 55 Replication 5 Simulation Time = 19. 28 Total Run time: 97. 3hrs As can be easily seen the scenario applied above has reduced the overall service time from 127. 46 to 97. 3 this indicates a difference of about 30 hours. The time taken to perform the entire work has reduced and customers are being served better and more effeciently. Queue at each work centre From the waiting time of entities at each work station table and graph above it can be easily seen that the highest average waiting time is 9. hrs as opposed to the highest from the as-is simulation which was 17 hrs. There is definitely an improvement within the system but we have not reached the optimum. From the figure and graph above it is seen that the admin officer still has the longest number of people waiting queue followed by the receptionist, the mechanics at the HGV repair at Bay 4 and then the mechanics at the Car Repair at Bay 4 respectively. We will seek to reduce this queue once more in order to effectively improve the performance of the system.

Resource utilization From the figure above, it can be detected that the highest utilised resources are mechanics 1, 2 then 3 what the experiment has done is to reduce the figures for both resources 2 and 3 but as we can see resource 1 is still high therefore further experiments will seek to reduce this. From the figure above, the three resources that deal with the most entities are the receptionist, the carwash machine and the admin officer. They are all dealing with all the entities that enter into the system.

Mechanics 4 & 5 treat the least number of entities and Mechanics 1, 2 and 3 have reduced drastically but there is a chance to reduce them further. What-if? Scenario 2 For the what-if? Scenario 2, a modification will be made to the 1st scenario. The initial changes made will remain as they are and then further changes will be made. The changes to be made to the model are: 3 different workers will be assigned to each bay, this means that Mecahnic 6 – Mechanic 12 will be created allowing each bay to be serviced by 3 individual mechanics.

As like the first scenario all the KPI’s are recalculated and shown below to record changes. Overall Service Time Replication 1 Simulation Time = 9. 81 Replication 2 Simulation Time = 12. 70 Replication 3 Simulation Time = 14. 26 Replication 4 Simulation Time = 10. 98 Replication 5 Simulation Time = 12. 84 Total Run time: 60. 59hrs As can be easily seen the scenario applied above has reduced the overall service time from 97. 3 to 60. 59 this indicates a difference of about 37 hours. The time taken to perform the entire work has reduced and customers are being served better and more effeciently.

Queue at each work centre What can be seen from the table and graph above is that only 3 resources have a high waiting time as opposed to 6 in the previous scenario the queues have reduced further by the application of this scenario to the model and entities are being served more quickly. The aim however is to minimize these queues to the barest minimum and since that have not been achieved another scenario might be necessary. From the table and graph above it is interesting to note that the number waiting for each work station has reduced but is still almost in the same proportion as the previous scenario.

Improvments would have to be made to this as well. From the graph above we notice that because the work has been spread across a wider array of mechanics the only resources that are being utilized the most are Mechanics 1, 2 and 9. A solution to this will be sought in the next scenario. As can be seen from the graph above, the number of entities that each resource treats has reduced drastically and this allows for better service to the entities. 4. 6 “SHOULD-BE” SCENARIO Considering all the scenarios treated a winning scenario has emerged and this will be proposed to the management of S ; J Garage for implementation.

This will lead to a maximization of all the key performance indicators and will benefit the company by increasing performance. For the “should-be” scenario, the model remains the same as the what-if? Scenario above but with these modifications: Each Bay is reduced to one workshop for all three vehicle types except bay 3 by removing 3 decide modules and leaving the last one only to account for the randomness at bay 3 for the vehicle type lorry. This will help the resources not to have to move around as much therefore saving time. KPI’s Overall Service Time Replication 1 Simulation Time = 9. 81 Replication 2 Simulation Time = 12. 70

Replication 3 Simulation Time = 14. 26 Replication 4 Simulation Time = 10. 98 Replication 5 Simulation Time = 12. 84 Total Run time: 60. 59hrs The overall simulation run time has remained constant and this is not bad because the overall run time is pretty low at this rate and need not be changed. Queue at each work station As can be deduced from the table and graph above, there is only 1 resource which reports a long queue and that is caused by the number of vehicles that have to be fixed at that bay. It is a queue that can be well managed and will not cause any disruptions to the system. The figures in this graph are as low as they can get.

The table and graph above indicate the drastic improvement in the number of people waiting to be served. The graph above shows the utilization of the resources in the model. Though not evenly spread it is the best result yet with 6 of the resources being utilized equally. The graph above shows the number of entities which are seized by each resource and this shows that each resource works on the right amount of resources. None of the resources are underutilized. The exceptionally high figures are attributed to the fact that the resources (receptionist, admin officer and car wash are required to treat every entity that comes into the system.

They are not required to carry out more than one task therefore it is alright for them to be high. 4. 7 RECOMMENDATIONS FOR FURTHER IMPROVEMENT Listed here are the recommendations for S & J Garage as regards their work flow processes. The 3rd scenario presented in this report has been labeled the “should-be” scenario. Therefore the changes modeled will form the basis for the recommendations. * The garage should consider hiring more workers to attend to the different kinds of faults. What this will invariably do is reduce the time wasted by the resources moving from one repair bay to the other.

It will also reduce the waiting time of the entities if their own resource is uniquely allocated to them. * Another major recommendation will be to dedicate only one garage to the repair of each fault as opposed to three and it will also address the same issue of the resources having to leave from one work station to the other. This will also hasten up the repair process as the time is the same for each vehicle and working together all 3 resources will be faster. * The last recommendation will be to keep the 3rd repair bay as it is. This is because of the machine breakdown which affects the timing of repair for Lorries in that bay.

An alternative would also be to invest in the repair of the machine and revert to the suggestion of the other 3 bays by making it one process instead of three. 5. 0 REFERENCES Al Bazi, A. and Dawood, N. (2011) ‘Improving The Performance And Reliability Of Construction Supply Chain Using Simulation: A Case Study For Doorsets Manufacturing’ -Doorssim-<http://moodle. coventry. ac. uk/ec/mod/resource/view. php? id=44745> Balci, O. (1990) ‘Guidelines for Successful Simulation Studies’. Proceedings of the 1990 Winter Simulation Conference (25-32),

Leave a Reply

Your email address will not be published. Required fields are marked *