Gis Network Analysis Eg

Table of Contents

Scenario 3

Introduction & Approach 3

What our group has described so far is an outline of our approach; the rest of the report will cover in greater detail what our geodatabase is about. 5

Assumptions 6

Limitations 7

Logical Model 9

Geodatabase 12

Network 13

Further developments 17

Conclusion 20

References 21

Annex 22

23

Scenario

Our group's problem statement is as follows: Larry, the vending-machine coin collector for the National University of Singapore (NUS) is frustrated with the high diesel prices. Larry then approached the department after he learnt that our department's Geographical Information Systems users have the capabilities to help him calculate the shortest route possible. Shamelessly, Larry then requested another feature, the fastest route possible by road, to be added for days when he is in a hurry. Thus our group was assigned to aid Larry in his quest to find the optimal path for cutting travel cost of both time and distance.

Introduction & Approach

It was obvious to our group that Larry requires a solution to his Vehicle Routing Problems (VRPs); he seeks to find a route or routes across a network for the collection of goods. From their earliest incarnation VRPs have been formulated as a distance or cost minimization problems (Clarke & Wright, 1964; Dazig & Ramser, 1959). This is very much still a problem today with "Nine out of ten research articles regarding route design in the context of transit routing written between 1967 and 1998 and reviewed by Chien and Yang (2000) employed a total cost minimization objective.

We identified that a geodatabase has to be first created before a network analysis can be made to solve Larry's problem, however, we decided take a more systematic approach; to follow most of the ten steps to designing geodatasbases to create our geodatabase.

The first four steps pertain to the conceptual design. The first would be to identify the information products that will be produced with our GIS. This, we mentioned earlier, is an analytical model of NUS road network capable of calculating the impedance for time and distance.

The second step is to indentify the key thematic layers based on our information requirements. We initially agreed that a base map of NUS, car parks, roads, vending machine locations and buildings should be in included as our key thematic layers. This was however; later shrunk to a base map of NUS, cap parks and roads, to be elaborated further in the report. The third step would be to specify the scale ranges and spatial representation for each thematic layer and the fourth step is to group representations into datasets.

The next three steps are part of our logical design, defining the tabular database structure and behavior for descriptive attributes; define the spatial properties of the datasets and to propose a geodatabase design. Please refer to the section titled, 'Logical Model' for more information.

For the physical design, the following steps were taken. First, we implemented a prototype; we then reviewed it and refined our design. Finally we made diagrams and this report to communicate our data model.

What our group has described so far is an outline of our approach; the rest of the report will cover in greater detail what our geodatabase is about.

Assumptions

To make an otherwise complicated and time-consuming task more manageable, our group decided to make a few assumptions. Firstly, we assumed that all roads in NUS are the same, be it the expressway or small narrow roads. Secondly, Larry will be travelling constantly at the maximum speed of 40km/hr within the campus. Thirdly, Larry will not be stopping at every single car park in NUS, but rather, car parks deemed to be the best choice for each part of NUS where Larry has to collect coins from will be selected for him. Next, in his course of travel to collect coins, Larry will not be stopping for any reasons other than the designated car parks (preferred car parks) he will be parking at, i.e. no traffic lights. And finally, our last assumption is that all turns made in our network will take an equal amount of time.

These assumptions naturally placed some limitations upon project, which will be discussed, in our next section. We do however believe that these assumptions will not affect the results severely as we are dealing with a relatively small area. If our project were dealing with a more complex network, like a big city, it would be more worthwhile and important to invest our resources to include the different road types and speed limits. As such, given the scale of the project and purpose, detailed investigations and pinpoint accuracy would be a waste of resources as compared to what we have proposed.

Limitations

Our group admits that there are several limitations in our product such as such as assumptions made, budget, time, human error, accuracy of data and access to data. In developing our product, several notions have to be assumed for easier management. One example would be that the speed of movement was assumed to be constant throughout the whole journey. This is clearly not true for most, if not all cases in reality. There are limitations in exactly reflecting the real world events, as there are unpredictable changes in the environment and scenarios. For example, weather, vehicle breakdown, and accidents, are something that cannot be accurately modeled both in the system and in the real world that we cannot factor out. Also, due to time constraints, most of the data that we modeled our product upon is based on secondary or preset data. These data, at best, gives only an estimation of the conditions that may affect the accuracy of out product. An example of this would be the use of "global turns" instead of first hand data on the turning times at various junctions.

However, these systemic and temporal constraints are