27.10.2014 Views

Form 10-K - Union Pacific

Form 10-K - Union Pacific

Form 10-K - Union Pacific

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

2011, effectively handling the 3% increase in carloads. Maintenance activities and weather disruptions,<br />

combined with higher volume levels, led to a 4% decrease in average train speed in 20<strong>10</strong> compared to a<br />

record set in 2009.<br />

Average Terminal Dwell Time – Average terminal dwell time is the average time that a rail car spends at<br />

our terminals. Lower average terminal dwell time improves asset utilization and service. Average terminal<br />

dwell time increased 3% in 2011 compared to 20<strong>10</strong>. Additional volume, weather challenges, track<br />

replacement programs, and a shift of traffic mix to more manifest shipments, which require additional<br />

terminal processing, all contributed to the increase. Average terminal dwell time increased 2% in 20<strong>10</strong><br />

compared to 2009, driven in part by our network plan to increase the length of numerous trains to improve<br />

overall efficiency, which resulted in higher terminal dwell time for some cars.<br />

Average Rail Car Inventory – Average rail car inventory is the daily average number of rail cars on our<br />

lines, including rail cars in storage. Lower average rail car inventory reduces congestion in our yards and<br />

sidings, which increases train speed, reduces average terminal dwell time, and improves rail car<br />

utilization. Average rail car inventory decreased slightly in 2011 compared to 20<strong>10</strong>, as we continued to<br />

adjust the size of our freight car fleet. Average rail car inventory decreased 3% in 20<strong>10</strong> compared to<br />

2009, while we handled a 13% increase in carloads during the period compared to 2009. We maintained<br />

more freight cars off-line and retired a number of old freight cars, which drove the decrease.<br />

Gross and Revenue Ton-Miles – Gross ton-miles are calculated by multiplying the weight of loaded and<br />

empty freight cars by the number of miles hauled. Revenue ton-miles are calculated by multiplying the<br />

weight of freight by the number of tariff miles. Gross and revenue-ton-miles increased 5% in 2011<br />

compared to 20<strong>10</strong>, driven by a 3% increase in carloads and mix changes to heavier commodity groups,<br />

notably a 5% increase in energy shipments. Gross and revenue-ton-miles increased <strong>10</strong>% and 9%,<br />

respectively, in 20<strong>10</strong> compared to 2009 due to a 13% increase in carloads. Commodity mix changes<br />

(notably automotive shipments) drove the variance in year-over-year growth between gross ton-miles,<br />

revenue ton-miles and carloads.<br />

Operating Ratio – Operating ratio is our operating expenses reflected as a percentage of operating<br />

revenue. Our operating ratio increased 0.1 points to 70.7% in 2011 versus 20<strong>10</strong>. Higher fuel prices,<br />

inflation and weather related costs, partially offset by core pricing gains and productivity initiatives, drove<br />

the increase. Our operating ratio improved 5.5 points to 70.6% in 20<strong>10</strong> and 1.3 points to 76.1% in 2009.<br />

Efficiently leveraging volume increases, core pricing gains, and productivity initiatives drove the<br />

improvement in 20<strong>10</strong> and more than offset the impact of higher fuel prices during the year.<br />

Employees – Employee levels were up 5% in 2011 versus 20<strong>10</strong>, driven by a 3% increase in volume<br />

levels, a higher number of trainmen, engineers, and yard employees receiving training during the year,<br />

and increased work on capital projects. Employee levels were down 1% in 20<strong>10</strong> compared to 2009<br />

despite a 13% increase in volume levels. We leveraged the additional volumes through network<br />

efficiencies and other productivity initiatives. In addition, we successfully managed the growth of our fulltime-equivalent<br />

train and engine force levels at a rate less than half of our carload growth in 20<strong>10</strong>. All<br />

other operating functions and support organizations reduced their full-time-equivalent force levels,<br />

benefiting from continued productivity initiatives.<br />

Customer Satisfaction Index – Our customer satisfaction survey asks customers to rate how satisfied they<br />

are with our performance over the last 12 months on a variety of attributes. A higher score indicates<br />

higher customer satisfaction. We believe that improvement in survey results in 2011 generally reflects<br />

customer recognition of our service quality supported by our capital investment program.<br />

Return on Average Common Shareholders’ Equity<br />

Millions, Except Percentages 2011 20<strong>10</strong> 2009<br />

Net income $ 3,292 $ 2,780 $ 1,890<br />

Average equity $ 18,171 $ 17,282 $ 16,058<br />

Return on average common shareholders' equity 18.1% 16.1% 11.8%<br />

33

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!