Best way to display date and time in a table with massive amounts of data?

Question

I have a table of massive amounts of data. Each table row is a time sensitive order that our users need to cover on a day to day basis. Order's appear from all around the country and we have orders going in and out of Canada and Mexico.

Currently date and time are displayed as either a single date and time or a date range.

10/31/2013 19:00

or

10/31/2013 19:00 - 11/2/2013 18:00

The table has two columns of date and time, one for pick up and one for delivery.

I want to make sure that we display the date and time in the most usable way possible for users to scan the table and be able to see the date and time and recognize it easily.

Some options that I have been thinking about are:

1) Oct 31 19:00
2) 10/31/2013 19:00
3) 10/31 19:00

Answer 1

I worked on a commercial product where this was an important issue. Here are a few considerations in addition to the article referenced above discussing localization formats:

How does the user identify the rows they need?

If they are trying to identify the orders by first-in-first-out you can consider changing the date to something like "2 hours ago" or "over 24 hours ago". This categorizes the orders into ones that need attention the fastest.

How much precision do the users need to identify a relevant row?

If the oldest order goes back a month, you probably don't need the year.

Can you combine two columns into one?

Does the date range mean something when there is a long delay between pick-up and delivery? How about when there is no delivery date a week after pick up? You could use natural language to describe what they're trying to identify by computing the difference between dates in their head. Additionally, if you are able to categorize the dates and leave the details to a drill-down, you can use a faceted search to avoid making them scan an entire table.

Note: Make sure that if you're hiding the precision the sorting still works with the highest precision available. Even if many items are "over 24 hours ago" it's common to expect the oldest is still last in the list.