{"id":7700,"date":"2020-09-27T17:50:27","date_gmt":"2020-09-27T15:50:27","guid":{"rendered":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/?p=7700"},"modified":"2020-09-27T17:50:27","modified_gmt":"2020-09-27T15:50:27","slug":"analytics","status":"publish","type":"post","link":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/2020\/09\/27\/analytics\/","title":{"rendered":"5 Steps to Get the Most From Your Analytics"},"content":{"rendered":"

Photo: Gettyimages.com\/enisaksoy<\/em><\/i><\/span><\/p>\n

You have to have put in place a way to measure the success of the analytics.<\/b><\/span><\/p>\n

Putting data to work has been a topic within the trucking industry for some time. As technologies that collect data have become more readily available and affordable, fleets of all sizes have begun studying the data from various sources within their business: operational data from their transportation management software, location and sensor data from telematics, vehicle data from the maintenance system, and other various sources such as safety and HR. Now-mandatory electronic logging devices record driver hours, vehicle location and engine operation, opening up a new trove of data.<\/span><\/p>\n

<\/div>\n

Much of that data can be used to gain businesses intelligence, or insights, into various parts of the business. However, you don\u2019t just turn on the data machine and have it spit out results. To be effective, any kind of effort should consider the following steps.<\/span><\/p>\n

1. Decide Where to Start<\/span><\/h2>\n

The reason companies take the effort to analyze their data is so they can better understand inefficiencies and other areas within the operation that can be improved.<\/span><\/p>\n

\u201cOften the list of problems is long, so there is a need for focus,\u201d explains Ashim Bose, chief data scientist and vice president of artificial intelligence\/machine learning at Omnitracs. \u201cPrioritize this list based on business value.\u201d<\/span><\/p>\n

Without a clear focus, he warns, it\u2019s easy to \u201cfall into the trap of \u2018boiling the ocean\u2019 in an unfocused manner.\u201d<\/span><\/p>\n

The list of things you could focus on is long: fuel economy, revenue per mile, routes, safety, etc. Each fleet will have its own priorities. \u201cEvery customer is different, but there are some common returns on investment from using the data,\u201d says Mix Telematics\u2019 Jonathan Bates, such as being able to get more affordable insurance rates by focusing on safety data and how to improve it.<\/span><\/p>\n

At Werner Enterprises, for instance, the focus is on equipment maintenance, safety, and utilization, according to Danny Lilley, vice president of fleet systems and technology and an HDT 2020 Truck Fleet Innovator.<\/span><\/p>\n

2. Get Your Data Ready<\/span><\/h2>\n

After settling on the problem or problems you want to investigate, you have to identify the type of data you need and where it can be found. That provides \u201ca starting point for data wrangling,\u201d Bose says.<\/span><\/p>\n

\u201cThe first thing I would do is decide what parts go together \u2014 what are apples, what are oranges,\u201d says Barry Brookins, director of data science at McLeod Software. For instance, if you are monitoring driver behavior, you don\u2019t want to compare a city driver with a long-haul, team driver.<\/span><\/p>\n

Chris Orban, vice president of data science at Trimble Transportation, agrees. \u201cOne of the first things that any fleet can do [is] understand the data you are looking at and understand what problem you are trying to solve.\u201d<\/span><\/p>\n

For instance, if you want to analyze vehicle failures, data sources might include engine data from a telematics system or data downloads, repair histories from maintenance software, warranty information, and so on. This information is most likely already being used by various people within your company, who Orban refers to as your \u201cbest data experts.\u201d<\/span><\/p>\n

Once you understand the data sources, then you move into getting \u201cclean data.\u201d<\/span><\/p>\n

This is a very important step, Lilley says. \u201cClean, organized data, so that the context is well understood, is essential to any analytical exercise and often the most time-consuming aspect of analytics.\u201d<\/span><\/p>\n

Criss Wilson, data scientist, McLeod Software, recommends recognizing what tools you already have. In other words, does the software you are using now have business intelligence capabilities? If yes, you next look at the different databases you have. \u201cIt\u2019s important to understand where the data is and how you are going to work with it,\u201d he says. This will include identifying the common data points between the databases. That might be vehicle ID, driver ID, order number, or other fields that are common between data sources.<\/span><\/p>\n

Then, it comes to cleaning the data. A common problem with getting your data in order is to get rid of duplicates, which are common when multiple people key in duplicate data, Orban says. For instance, you may have multiple data on one driver or multiple customer entries.<\/span><\/p>\n

Cleaning data can be time-consuming. Orban recommends that fleets work collaboratively with their software vendors on this task. While many vendors can \u201cclean\u201d the data, there may be reasons for some duplicates that they wouldn\u2019t know about, which is why it\u2019s important to have the fleet\u2019s data experts involved.<\/span><\/p>\n

Focusing on processes helps ensure the data going in is useful later. Various departments are all contributing to databases on vehicles, drivers, customers, freight lanes, etc. \u201cYou have to make sure everyone is speaking the same \u2018data language,\u2019\u201d Orban adds. For instance, there should be clear definitions for things such as reportable accidents or late loads, so the data input is consistent across staff and departments.<\/span><\/p>\n

3. Have a Plan<\/span><\/h2>\n

Before jumping into the deep end of the data pool, understand what you are looking for. \u201cIt is very important to have a hypothesis that can be tested to validate the result,\u201d Lilley says. Each of these tests can show whether or not your approach can provide the end result you are looking for. \u201cContinuous experimentation is critical to identifying the things that could be extremely important.\u201d<\/span><\/p>\n

You have to have put in place a way to measure the success of the analytics, says Mike Branch, vice president of data and analytics at Geotab.<\/span><\/p>\n

In addition, you need to \u201cestablish the time frame,\u201d Trimble\u2019s Orban advises. If you focus on too short a time frame, the past week for instance, what you find may not be a trend, but just a seasonal variation. Therefore, it\u2019s key to look at a long enough time period to tease out simple variations.<\/span><\/p>\n

4. Let the Data Do the Talking \u2013 Keep an Open Mind<\/span><\/h2>\n

When the analytics process starts producing results, it\u2019s important to keep an eye on the data you are working with, \u201ceverything from data quality to volume,\u201d Branch says. In the initial phases, it may help to develop \u201csimple visualizations via graphs that allow both the business and data experts to be aligned.\u201d<\/span><\/p>\n

Most fleet managers already have a fairly good understanding of their company. But sometimes, the data results may not jibe with their expectations. That doesn\u2019t mean that either of the two is wrong.<\/span><\/p>\n

\u201cIf the data doesn\u2019t agree with what you\u2019ve been finding, that can be tough,\u201d says McLeod\u2019s Brookins. For instance, if the analytics ranks a customer higher on the list than the manager would have, look at other factors the analytics model considered \u2013 or didn\u2019t consider. Maybe drivers score that customer tops because the people there are great to work with and get them in and out on time. If you only look at rates in your data analysis, you might overlook other factors that make a customer a good customer. Brookins says that kind of 360-degree view can help you make better choices.<\/span><\/p>\n

Even if you already have a pretty good handle on the metrics you use to measure your business, the bottom line is to keep an open mind. Understand that there is no perfect analysis, so don\u2019t blindly accept a model\u2019s results. Be open to learning something new.<\/span><\/p>\n

\u201cYou never know where the data will lead, and often you stumble upon an insight that you were not considering,\u201d Lilley says.<\/span><\/p>\n

5. Put What You\u2019ve Learned to Work<\/span><\/h2>\n

Gaining new insight into how your company is operating is not the end the of the journey. Putting those insights to work is the payoff. \u201cTeams often focus on generating insight, but sometimes neglect to focus on how those insights will be utilized,\u201d Lilley says.<\/span><\/p>\n

You can\u2019t get bogged down in the data. \u201cYou can have a problem with trying to look at too many pieces of data,\u201d Brookins says.<\/span><\/p>\n

But looking at too little can also be a problem. For instance, ranking drivers by revenue generated on one list and safety on another list may result in completely different lists. A deep analysis of data, on the other hand, can combine those data points, plus others, to provide a more accurate picture.<\/span><\/p>\n

Just as important as finding the insights is getting that information to the right people on your staff. Branch says that could be accomplished with an email alert, a dashboard or a software application. But whatever method you use, it needs to \u201cconvey the result of the project to the people that require the insight in the business.\u201d<\/span><\/p>\n

Perhaps more important, Branch says, is \u201ctracking the performance of the insight that you\u2019ve developed against the success criteria outlined\u201d earlier in the process.<\/span><\/p>\n

Lilley says that in the future, advances in business intelligence and machine learning technologies could automatically integrate the insights gained into your business processes. In that case, \u201cthe analytics will be performed by technology,\u201d which will then initiate actions, providing almost immediate operational adjustments. In this scenario, the data knows<\/span> best.<\/p>\n

 <\/p>\n

by Jim Beach<\/a><\/p>\n

Source: https:\/\/www.truckinginfo.com<\/a><\/span><\/p>\n

CUT COTS OF THE FLEET WITH OUR AUDIT PROGRAM<\/strong><\/a><\/h3>\n

\"\"<\/a><\/p>\n

The audit is a key tool to know the overall status and provide the analysis, the assessment, the advice, the suggestions and the actions to take in order to cut costs and increase the efficiency and efficacy of the fleet. We propose the following fleet management audit.<\/p>\n

FLEET MANAGEMENT AUDIT<\/strong><\/a><\/h3>\n","protected":false},"excerpt":{"rendered":"

Photo: Gettyimages.com\/enisaksoy You have to have put in place a way to measure the success of the analytics. Putting data to work has been a topic within the trucking industry for some time. As technologies that collect data have become more readily available and affordable, fleets of all sizes have begun studying the data from…<\/p>\n","protected":false},"author":3,"featured_media":7701,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[18],"tags":[39],"_links":{"self":[{"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/posts\/7700"}],"collection":[{"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/comments?post=7700"}],"version-history":[{"count":1,"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/posts\/7700\/revisions"}],"predecessor-version":[{"id":7702,"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/posts\/7700\/revisions\/7702"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/media\/7701"}],"wp:attachment":[{"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/media?parent=7700"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/categories?post=7700"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/tags?post=7700"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}