Regression Analysis

Introduction

Our regression analysis was done on OMNITRANS fuel consumption. This has been an ongoing issue for OMNITRANS where there seems to be an inconsistency with there CNG fuel consumption. There continues to be variance in what is consumed each day compared to the amount of miles driven. This issue is very important to OMNITRANS because it makes it very difficult to plan for future use with the CNG industry. OMNITRANS wants to have a consistency with CNG use so they can plan for budgeting purposes and new contracts that are connected with the CNG usage. We are going to establish what we believe will be a good analysis for OMNITRANS to look at and establish what is needed in order to improve there fuel consumption within the industry. This will give them an overall picture of where there is a correlation between the two factors. In our particular study it is fuel consumed and miles driven.

Data Researched

What we have done is taken 5 days worth of data points from the Fleetwatch system's reports. Fleetwatch is based for all recording purposes for all of the OMNITRANS facility. This particular report gives full details of how much fuel is consumed on each day based on filling station onsite at the OMNITRANS facility. We took every 5th days worth of data from the first on. That data point days were the 1st, 6th, 11th, 16th and the 21st. Each of these days that was chosen also represented a day where all buses were in full use with little to no difference in there operating schedule. This was established as the best way of getting a good variation of data.

We took the data from both filling stations to come up with our X factor for each days worth of data points. This gave us an assortment of total fuel consumed for each day. That way we are not going to have an exact same number for each day. This represents a better view of the fuel used. Instead of taking an average of all fuel consumed for a month by using the 5 different days we are able to look at a better sample of what is actually happening with fuel consumption.

Fuel Consumed Miles Driven

When X Y

Day1 2904.2 8712.6

Day6 2807.1 8421.3

Day11 4128.3 12384.9

Day16 3489.3 10467.9

Day21 2424.3 7272.9

Coefficient of Determine = 1

Fuel Consumed Miles Driven

2500 7500

3500 10500

Next was the issue of establishing the Y factor. With this we ran into an issue of how many miles for each day. Even though OMNITRANS has set schedules for there buses there seemed to be an issue of establishing a breakdown of each bus bases on the different routes. Some buses drive only10 miles a day while another bus may drive up to 400 in single work day. What we have done is taken the average miles per gallon and multiplied by the amount of fuel consumed in gallons. We found that this is the best way of showing the correlation between the X and Y.

The opportunity for this is that if we can establish a norm of what should be happening with fuel consumption we would have a better way of establishing what is going on with the OMNITRANS buses. Below you will find our regression analysis for the data that was explained between the X and Y.

Interpret Results

By the data above it will show that there is a direct relationship between fuel consumed and miles driven. As fuel consumed is increase it will increase miles driven. In average the buses consume one gallon of gas for every 3 miles. On the other hand some buses may consume more if there is a problem with the engine or change in weather.

From the graph it will show a correlation between fuel consumed and miles driven, with the correlation coefficient being equal to 1. This indicates that the number of fuel consumed in a perfect positive linear with miles driven. Also from the graph all points do not lay exactly on the line, if they did we would predict with 100 percent accuracy.

State the limitations of the analysis

With this analysis there are some limitations such as something interferes with the routine for that day. Such as engine failure, if the engine is starting to fail it will start to work harder when in use and this will lead to less miles per gallon of gas. Miles driven may also change when the weather changes. For example, when it is summer they will need to run the air conditioner and this will result in more gas used per mile. Any of these examples or other will lead for the results of the analysis to change and when these figures are not included in the analysis it can the state the limitations.

The Significance of the Results to the Organization as a Whole

Primarily the results will impact the organization as a whole financially. Nearly 20% of the world's primary energy demand is for transport fuels, the vast majority of which is satisfied by oil based products. This represents a great opportunity for gas, which could power the road transport sector firstly as Compressed Natural Gas (CNG) or Liquefied Petroleum Gas (LPG) in

specific applications such as buses, taxis and hybrid vehicles. This market is expected to grow, particularly in urban areas where the environmental benefits of gas-based products are important for air quality. Later gas could be used as a primary feedstock for producing hydrogen to fuel cells.

What does this all mean for gas consumption? Research has shown that new distributed and merchant generating capacity is anticipated to be predominantly gas fired. Large non-central generating capacity is also predominantly gas fired today and is expected to remain so in the future. While gas consumption in the divested capacity may fall somewhat over the next 5 to 10 years as the units are repowered or retired by merchant generators, that decline will be offset by growth from new gas-fired merchant capacity. However, the decline in gas consumption will vary significantly by region depending on the extent of gas use today, merchant activity, and regional fuel preference.

What are Your Alternative Measures to the Problem?

Buses consume more than twice the amount of fuel used by the