The Information, Tours and Tickets (ITT) office could use a system to assist them in creating trip schedules. In this paper I will outline a plan for a Decision Support System (DSS) that will assist ITT in creating schedules for their tours. This system will also track customer surveys and hold data about all of ITTs trips. They already have some computer systems, a spread sheet program and a data base management system (DBMS) which can all be used to build a small DSS. Using the DBMS and the spread sheet software I have designed a system to assist them in making decisions about scheduling trips. This system also allows them access to information about all of ITTs trips and the feedback from customers about each trip. In the next few paragraphs I go through the major steps in developing a system of this nature. A system of this type goes through several phases in its development. I start with the planning phase and go on to discuss research, analysis and conceptual design. Then I talk a little about the models used in this system. I finish by talking about the actual design, the construction, and the implementation of the new ITT system. I finish the paper with a discussion of maintaining the system.
The first step in building any DSS is planning. Planning is basically defining the problem. The planning also involves an assessment of exactly what is needed. In this case I deal with trip scheduling. In the case description this would include: How many trips to offer, the days of the week to have particular trips, and when to cancel trips. Obviously the scheduling ties to other information such as profit and participation, but for this paper I will only cover the scheduling portion of ITTs problem. Therefore I have defined the problem as a basic scheduling problem. I see a need for ITT to better schedule trips using the information they have and using the information they collect from customer surveys. With the problem defined we can now look at what information is needed to further analyze the problem.
After a problem is defined, information must be collected. The research phase of system development is just that, collecting information. The information collected will be used in the next phase of development to further analyze the problem and it will be used in this case to build the databases. The databases will then be used with decision support system (DSS) models to assist ITT in making scheduling decisions. Information in this case can come from their current schedules and trip description fliers. Also during this stage of development the current resources are assessed. This would include ITTs current information systems and their current budget. And, information such as Navy or ITT policies are collected as a reference. Once all the information is collected than the system can move to the next stage of development, analysis.
With all the data and information collected analysis of it begins. In this stage we determine what needs to be done to solve the problem. No work, on a new system, is started yet, but a system is conceptualised and possible solutions are identified. Also in this stage a final solution to the problem is chosen and system passes through another stage of development. For the ITT problem I have chosen a simple Management Information System (MIS) with small decision support models to aid in creating schedules. This system will provide ITT with the information they need to make decisions about scheduling their trips as well as allow them to create the schedules directly from computer models. I will discuss the models in the next paragraph. The system would not draw conclusions, but simply show the pros and cons to certain choices. The MIS portion of the system will simply provide information to the users and to the DSS. The DSS portion of the system will allow a schedule to be created using resources in an optimum manner. I decided to go with a small and simple system because of ITTs limited resources and because of a high employee turn-over. A complicated system would not be feasible in such an environment where new employees are constantly having to be trained to use it.
In this paragraph I side step from the development a little to talk about the models used in the system. As stated earlier the models used in this system should be kept simple and small if possible. Using standard spread sheet software, models can be created that will show the optimal schedule for trips. The basic information required for these models should include bus schedules, reservation requirements, customer satisfaction information, and cost data. Other data can also be added to assist in decisions. The models would first approximate the participation for each proposed trip. Then another model would determine if the trip is feasible given the costs involved. The next model would determine if the trip is even possible considering what is required as far as reservations and transportation. Another model could also determine if the trip would be able satisfy the customers given the past customer inputs. Finally after determining whether each trip is worth offering the ITT employees could use the computer to generate a new schedule. Now that we know what the system should do we can turn our attention to the design of the system.
In the design phase of DSS development the new system is designed to solve the problem. Here the information collected and the resources identified are examined to decide exactly what must be done and in what manner to solve the problem. Diagrams may be drawn to show how the components will fit together. Also the ground work is laid for the construction phase. In the ITT case I designed a database that contains information on all their trips along with information obtained from the customer surveys. This information is then combined with bus and reservation data in a standard spread sheet where it is manipulated to optimize the trip schedule. A manager or an employee can then use the information and the data in the database to create a calendar of events using an inexpensive program 'Calendar Creator.' Which is what ITT currently uses to create schedules manually.
The construction of a system is the bringing together of all the required parts and making the system do what it's supposed to do. In this case the system I have designed will only require a minimum of additional resources. I designed the system to work on their existing computers using their existing software. The databases and the spreadsheet models could be built by knowledgeable employees with minimal outside help. Once constructed the system could be run and results compared with the old system to determine if it is functioning properly. The results could also tell if the system is optimizing the schedule or just speeding up what is already done manually.
Implementing any system is the process of putting it into use. In this project the implementation phase should be a fairly easy conversion. The old way of manually deciding on trips and putting the results into Calendar Creator is simply replaced with an automated selection of trips that the employee can use to create a calendar. When the system is operating normally it should improve the way ITT does business.
After implementing this system it will have to be maintained. New models will have to be added and old models will have to be changed or removed. With the simple models used in this system that should not be difficult. The hardest component in this system to maintain will be the databases. They will have to contain the most current data in order for the system to operate properly. I suggest a data checking module be added at some point in order to maintain data consistency. Inconsistent data is something that can degrade the system performance and cause it to give inaccurate or incorrect information. A data checking module will insure that the information entered into the system is accurate and consistent with the rest of the system.
There could be many solutions to this problem, but given the limited budget most are not feasible. The system I have designed should be more than sufficient to assist them in creating schedules faster and more efficiently. Also it will give customers more of what they want and should improve repeat business.