<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Joe Chop on Medium]]></title>
        <description><![CDATA[Stories by Joe Chop on Medium]]></description>
        <link>https://medium.com/@joe.chop?source=rss-dfdc2f4de04f------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*n8BiMlIvkeHSaOZD</url>
            <title>Stories by Joe Chop on Medium</title>
            <link>https://medium.com/@joe.chop?source=rss-dfdc2f4de04f------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 06:33:34 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@joe.chop/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Exploring Visualizations — Area Charts, Regions Maps, and Markdown]]></title>
            <link>https://medium.com/statengine/exploring-visualizations-area-charts-regions-maps-and-markdown-89df617d9a8b?source=rss-dfdc2f4de04f------2</link>
            <guid isPermaLink="false">https://medium.com/p/89df617d9a8b</guid>
            <category><![CDATA[nfors]]></category>
            <category><![CDATA[data-visualization]]></category>
            <dc:creator><![CDATA[Joe Chop]]></dc:creator>
            <pubDate>Sat, 31 Aug 2019 00:45:59 GMT</pubDate>
            <atom:updated>2019-08-31T00:45:59.097Z</atom:updated>
            <content:encoded><![CDATA[<h3>Exploring Visualizations — Area Charts, Regions Maps, and Markdown</h3><p>The most commonly used visualizations in NFORS Analytics, if not all data science, are pie, bar, table, metric, and gauge visualizations. But did you know there 20 different visualizations types in NFORS Analytics? Today we’re going to explore some of the less common visualizations as we work through examples and discuss possible use cases.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zSkqL56yHSGtyxmMmGSPyg.png" /><figcaption>NFORS Analytics Visualization</figcaption></figure><h3><strong>Area Chart</strong></h3><p>I like to think of an area chart as the combination of a line and pie chart. A line chart typically shows a variable over time, for example, incident count over time. A pie chart typically shows composition of a variable, for example, the breakdown of incident categories. An area chart combines these two concepts, to visualize both dimensions, for example, incident categories over time. An area chart is visually similar to a line chart, but the area between the line and the x-axis is also filled with colors or patterns showing the composition of the composition variable.</p><p>Let’s try building an area chart to visualize incident categories over time:</p><ol><li>Navigate to the <strong><em>Visualize</em></strong> Tab</li><li>Click <strong><em>+</em></strong></li><li>Click <strong><em>Area</em></strong></li><li>Choose the incident index</li><li>Adjust the time range to a larger time period, such as<strong><em> Last 2 years</em></strong></li><li>Under Buckets, click <strong><em>X-axis</em></strong></li><li>Select <strong><em>Date Histogram</em></strong></li><li>Press the play button</li><li>Click <strong><em>Add sub-buckets</em></strong></li><li>Click <strong><em>Split Series</em></strong></li><li>Choose <strong><em>Terms</em></strong></li><li>Under Field, Type/Select <strong><em>description.category</em></strong></li><li>Press the play button</li><li>Save your work if desired</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-S8nUClFGCMByDxL10OFfw.gif" /><figcaption>Building an Area Chart</figcaption></figure><h4><strong>Further Exercises</strong></h4><p>With over 100 fields collected in NFORS Analytics, your possibilities for area charts are endless, but here are a few suggestions to practice:</p><ul><li>Incident Count By Organization (station, battalion, district, etc)</li><li>Incident Count By Shift</li><li>Turnout Times</li></ul><p><strong>Pro-tip</strong>: For the composition variable, you generally want to stick with a field with no more than 5 categories for legibility.</p><h3><strong>Region</strong> <strong>Map</strong></h3><p>Region maps, also known as choropleth maps, are similar to coordinate maps, in which they visualize the geospatial dimension of data. However, instead of just plotting the data on map, region maps join the data with boundary vector shapes. The vector shapes are colored using a gradient, with higher intensity colors indicating larger values, and lower intensity colors indicating smaller values. Region maps are valuable when you want to visualize not only the data, but the surrounding geospatial context, such as the zip code, jurisdiction boundaries, etc.</p><p>Let’s try building an region map, showing incidents by zip code:</p><ol><li>Navigate to the <strong><em>Visualize</em></strong> Tab</li><li>Click <strong><em>+</em></strong></li><li>Click <strong><em>Region Map</em></strong></li><li>Choose the incident index</li><li>Adjust the time range to a larger time period, such as<strong><em> Last 2 years</em></strong></li><li>Under Buckets, click <strong><em>shape field</em></strong></li><li>Select <strong><em>Terms</em></strong></li><li>Under field, type <strong><em>address.postal_code</em></strong></li><li>Change size to 100 (this number should be larger than the number of zip codes in your jurisdiction)</li><li>Switch to <strong><em>Options</em></strong> tab</li><li>Select USA Zip Codes as the <strong>Vector Layer</strong></li><li>Unclick <strong><em>Display warnings.</em></strong></li><li>Press the play button</li><li>Save your work if desired</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wcUUkIlcKAs5DuzhXvQe2g.gif" /><figcaption>Building a Region Map</figcaption></figure><p><strong>Further Exercises</strong></p><p>NFORS currently has zip codes, as well as 2010 Census Tracts vector layers for each state available for use. Try building some of these visualizations based on those layers:</p><ul><li>Incident Count By Census Tract</li><li>Average Turnout Time By Census Tract</li><li>Total Commitment/Utilization By Zip Code</li></ul><p><strong>Coming-Soon</strong>: Look forward to a new feature allowing users to upload your own vector layers, such as station jurisdictions. This feature is slated for release 3rd Quarter 2019.</p><h3><strong>Markdown</strong></h3><p>The Markdown visualization is unique in that it doesn’t actually display fire data, but it is a useful visualization to further enhance dashboards and “tell your story”. <a href="https://en.wikipedia.org/wiki/Markdown">Markdown</a> is a lightweight computer language, that allows you to quickly add structured text and links to your dashboards. Markdown leverages syntax and special characters such as *, to format the surrounding the text. There is a little bit of a learning curve if you want to take full advantage of all the syntax rules, but the beauty of markdown is you can just start typing as normal if formatting is not critical.</p><p>The full guide for markdown syntax is here: <a href="https://help.github.com/en/articles/basic-writing-and-formatting-syntax">https://help.github.com/en/articles/basic-writing-and-formatting-syntax</a></p><p>As a quick example, lets try experimenting with some markdown</p><ol><li>Navigate to the Visualize Tab</li><li>Click <strong><em>+</em></strong></li><li>Click Markdown</li><li>Copy and paste the following:</li></ol><blockquote># This is a heading</blockquote><blockquote>## This is another heading</blockquote><blockquote>I’m **bold**</blockquote><blockquote>I’m *italicized*</blockquote><blockquote>&gt; I’m a quote</blockquote><blockquote>Find out more here: [IPSDI](<a href="https://i-psdi.org/">https://i-psdi.org/</a>)</blockquote><blockquote>I can do lists too:<br>- 1<br>- 2 <br>- 3</blockquote><p>5. Press play</p><p>6. Experiment with the sytnax in the editor, experimenting with the markdown syntax</p><p>7. Press play</p><p>8. Save your work if desired</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8HiWqPrNeZEvWXx6eM_Vnw.gif" /><figcaption>Building a Markdown</figcaption></figure><h4>Further Exercises</h4><p>Try creating a few more text areas for your dashboards:</p><ul><li>Provide definitions of metrics in terminology more familiar with your department</li><li>Provide detailed textual analysis, commentary, or additional explanation of a chart or dashboard</li><li>Provide instructions for users (how to use filters, who to contact if you need help, etc).</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=89df617d9a8b" width="1" height="1" alt=""><hr><p><a href="https://medium.com/statengine/exploring-visualizations-area-charts-regions-maps-and-markdown-89df617d9a8b">Exploring Visualizations — Area Charts, Regions Maps, and Markdown</a> was originally published in <a href="https://medium.com/statengine">statengine</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Organizing with Workspaces]]></title>
            <link>https://medium.com/statengine/organizing-with-workspaces-2eda708d295a?source=rss-dfdc2f4de04f------2</link>
            <guid isPermaLink="false">https://medium.com/p/2eda708d295a</guid>
            <category><![CDATA[nfors]]></category>
            <category><![CDATA[fire-data]]></category>
            <category><![CDATA[data-analytics]]></category>
            <category><![CDATA[public-safety]]></category>
            <dc:creator><![CDATA[Joe Chop]]></dc:creator>
            <pubDate>Fri, 31 May 2019 22:12:20 GMT</pubDate>
            <atom:updated>2019-05-31T23:45:41.882Z</atom:updated>
            <content:encoded><![CDATA[<p>We’re excited to rollout a new feature today called, <strong><em>workspaces!</em></strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*F-6dfLBBHuFjvMOeuWCeBw.png" /><figcaption>Workspaces</figcaption></figure><p>The concept of workspaces originally came from a request from one of our larger metropolitan departments. They had several different users groups, including data analysts, operations/performance management, and a multi-level hierarchy of battalion, district, and station leadership users. As each user group has a different role in the organization, you might imagine the dashboards they were creating were drastically different both in structure and underlying theme. One dashboard was for station turnout times, one for end of the year reporting, one for municipality reports, one for station location analysis, and the list goes on. Managing all these dashboards, and more importantly, user access — or who can modify which dashboards — soon become problematic.</p><p>With <strong><em>workspaces</em></strong>, you can now organize your dashboards, visualizations, and saved searches into meaningful categories such as Operations, Accreditation, Battalion, or Station Dashboards. Once inside a workspace, you will only have access to the dashboards and visualizations that have been created in that workspace. When creating a workspace, you can also grant other users read/write access, or keep the workspace only visible to yourself. This allows you to have your own private playground for example, if you didn’t want to destroy the “gold” copy.</p><h3>Lets get started!</h3><p>Login to NFORS and click Dashboard on the left-hand menu. Instead of directly entering the dashboard arena, you will now be presented with a workspaces screen:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*e82BnaRzvCfptzk3U50wvQ.png" /></figure><h4>Note: You will notice a “default” workspace. For existing users, all your existing dashboards have been migrated to this workspace. Feel free to continue using this workspace, and start creating new workspaces when you’re ready or your use cases grow.</h4><p><strong>Creating a new workspace</strong></p><ol><li>From the workspaces landing page, click Create New</li><li>Fill out the general details</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZRWgHALBcUBO0g4yxqpmTQ.png" /><figcaption>General Details</figcaption></figure><p><strong>Name</strong>: An identifying name for the workspace</p><p><strong>Description</strong>: A text area to describe the intention of the workspace</p><p><strong>Color</strong>: A color can be selected to quickly identify the workspace</p><p><strong>Create Sample Dashboards</strong>: If <strong>YES</strong>, we will load the workspace with our default or “stock” Incident and Responses Dashboards. Toggle this button, if you would like an empty workspace.</p><p>3. Assign user permissions</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*LYawD8Tg4UxGiA6X8nTukQ.png" /><figcaption>User Permissions</figcaption></figure><p>When you create a workspace, by default, it will be private. However, if you are creating a shared workspace, you will want to adjust ownership and access permissions.</p><p>To assign another user as a co-owner, click the dropdown in the Owner column on the appropriate row.</p><ul><li><strong>YES</strong> will make the user a co-owner. This gives the user <strong>WRITE </strong>access to the workspace and allows the user to grant or revoke permissions, as well as delete a workspace.</li><li><strong>NO</strong> will remove the user as a co-owner. The user will maintain any access permissions they have.</li></ul><p>Generally, we only recommend a limited amount of co-owners (just yourself and a backup). For other members of your workspace that are not co-owners, you’ll want to assign the appropriate permissions via the dropdown in the Access Permission column.</p><ul><li><strong>READ</strong> will give the user read access to a workspace’s saved objects. This allows the user to filter, and explore dashboards and visualizations. However, the user will <strong>NOT</strong> be able to create or modify any dashboards or visualizations in this workspace.</li><li><strong>WRITE</strong> will give the user write access to a workspace’s saved objects. This allows the user to create, edit, and delete dashboards and visualizations.</li><li><strong>NONE</strong> will remove the user’s access to the workspace. No saved objects will be deleted.</li></ul><p>4. Click Create</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3kSolbfDocO5y16SveOUcw.png" /><figcaption>New Accreditation Workspace</figcaption></figure><p>5. Enter your new workspace by clicking the new card shown.</p><p><strong>Editing a workspace</strong></p><p>Owners can edit user permissions and general workspace details at anytime, via the dropdown in the right hand corner of the workspace card.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*etAG_XDiRm-HwgUiTADrRQ.png" /></figure><p><strong>Deleting a workspace</strong></p><p>Owners can delete a workspace, via the dropdown in the right hand corner of the workspace card. <strong>WARNING: this will delete all the saved objects in the workspace!</strong></p><p><strong>Copy data between workspaces</strong></p><p>Say you’ve got some sweet looking visualizations you want copied over to a new workspace. No problem! We got you covered. You’ll need to export all the objects from your old workspace and import them to your new workspace.</p><p><strong>Exporting Objects</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8ZWIJvbHWxpyIp6ii5sFgA.gif" /><figcaption>Exporting Objects</figcaption></figure><ol><li>Enter the “old” workspace, or the workspace that you would like to copy the objects from.</li><li>Click the Management Tab</li><li>Click the Saved Objects Option</li><li>Click the checkbox to select all objects, or selectively select the subset of objects you would like to migrate over</li><li>Click Export and Confirm</li><li>A .json file should be downloaded to you computer</li></ol><p><strong>Importing Objects</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*DDas-Vu7gxXLDaBy1d07nQ.gif" /><figcaption>Importing Objects</figcaption></figure><ol><li>Enter the “new” workspace, or the workspace that you would like to copy the objects into to.</li><li>Click the Management Tab</li><li>Click the Saved Objects Option</li><li>Click Import, and select the .json file generated in the export</li><li>Patiently wait a few moments!</li><li>Once complete, press Done</li></ol><p><strong><em>We hope you enjoy the new workspaces feature! Stay tuned for an exciting new feature we are working on to selectively pre-populate workspaces with even more “stock” dashboards including accreditation tables, station location analysis, station turnout dashboards, and more.</em></strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2eda708d295a" width="1" height="1" alt=""><hr><p><a href="https://medium.com/statengine/organizing-with-workspaces-2eda708d295a">Organizing with Workspaces</a> was originally published in <a href="https://medium.com/statengine">statengine</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Organizing with Workspaces]]></title>
            <link>https://medium.com/@joe.chop/organizing-with-workspaces-f6cc186a5ca5?source=rss-dfdc2f4de04f------2</link>
            <guid isPermaLink="false">https://medium.com/p/f6cc186a5ca5</guid>
            <category><![CDATA[fire-department]]></category>
            <category><![CDATA[nfors]]></category>
            <category><![CDATA[data-analytics]]></category>
            <category><![CDATA[workspace]]></category>
            <dc:creator><![CDATA[Joe Chop]]></dc:creator>
            <pubDate>Fri, 31 May 2019 22:09:56 GMT</pubDate>
            <atom:updated>2019-05-31T22:09:56.110Z</atom:updated>
            <content:encoded><![CDATA[<p>We’re excited to rollout a new feature today called, <strong><em>workspaces!</em></strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*F-6dfLBBHuFjvMOeuWCeBw.png" /><figcaption>Workspaces</figcaption></figure><p>The concept of workspaces originally came from a request from one of our larger metropolitan departments. They had several different users groups, including data analysts, operations/performance management, and a multi-level hierarchy of battalion, district, and station leadership users. As each user group has a different role in the organization, you might imagine the dashboards they were creating were drastically different both in structure and underlying theme. One dashboard was for station turnout times, one for a end of the year reporting, one for municipality reports, one for station location analysis, and the list goes on. Managing all these dashboards, and more importantly, user access — or who can modify which dashboards — soon become problematic.</p><p>With <strong><em>workspaces</em></strong>, you can now organize your dashboards, visualizations, and saved searches into meaningful categories such as Operations, Accreditation, Battalion, or Station Dashboards. Once inside a workspace, you will only have access to the dashboards and visualizations that have been created in that workspace. When creating a workspace, you can also grant other users read/write access, or keep the workspace only visible to yourself. This now allows you to have your own private playground for example, if you didn’t want to destroy the “gold” copy.</p><h3><strong>Lets get started!</strong></h3><p>Login to NFORS and click Dashboard on the left-hand menu. Instead of directly entering the dashboard arena, you will now be presented with a workspaces screen:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*e82BnaRzvCfptzk3U50wvQ.png" /></figure><h4>Note: You will notice a “default” workspace. <strong>For existing users, all your existing dashboards have been migrated to this workspace. Feel free to continue using this workspace, and start creating new workspaces when your ready or your use cases grow.</strong></h4><p><strong>Creating a new workspace</strong></p><ol><li>From the workspaces landing page, click Create New</li><li>Fill out the general details</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZRWgHALBcUBO0g4yxqpmTQ.png" /><figcaption>General Details</figcaption></figure><p><strong>Name</strong>: An identifying name for the workspace</p><p><strong>Description</strong>: A text area to describe the intention of the workspace</p><p><strong>Color</strong>: A color can be selected to quickly identify the workspace</p><p><strong>Create Sample Dashboards</strong>: If <strong>YES</strong>, we will load the workspace with our default or “stock” Incident and Responses Dashboards. Toggle this button, if you would like an empty workspace.</p><p>3. Assign user permissions</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*LYawD8Tg4UxGiA6X8nTukQ.png" /><figcaption>User Permissions</figcaption></figure><p>When you create a workspace, by default, it will be private. However, if you are creating a shared workspace, you will want to adjust ownership and access permissions.</p><p>To assign another user as a co-owner, click the dropdown in the Owner column on the appropriate row.</p><ul><li><strong>YES</strong> will make the user a co-owner. This gives the user <strong>WRITE </strong>access to the workspace and allows the user to grant or revoke permissions, as well as delete a workspace.</li><li><strong>NO</strong> will remove the user as a co-owner. The user will maintain any access permissions they have.</li></ul><p>Generally, we only recommend a limited amount of co-owners (just yourself and a backup). For other members of your shared area that are not co-owners, you’ll want to assign the appropriate permissions via the dropdown in the Access Permission column.</p><ul><li><strong>READ</strong> will give the user read access to a workspace’s saved objects. This allows the user to filter, and explore dashboards and visualizations. However, the user will <strong>NOT</strong> be able to create or modify any dashboards or visualizations in this workspace.</li><li><strong>WRITE</strong> will give the user write access to a workspace’s saved objects. This allows the user to create, edit, and delete dashboards and visualizations.</li><li><strong>NONE</strong> will remove the user’s access to the workspace. No saved objects will be deleted.</li></ul><p>4. Click Create</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3kSolbfDocO5y16SveOUcw.png" /><figcaption>New Accreditation Workspace</figcaption></figure><p>5. Enter your new workspace by clicking the new card shown.</p><p><strong>Editing a workspace</strong></p><p>Owners can edit user permissions and general workspace details at anytime, via the dropdown in the right hand corner of the workspace card.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*etAG_XDiRm-HwgUiTADrRQ.png" /></figure><p><strong>Deleting a workspace</strong></p><p>Owners can delete a workspace, via the dropdown in the right hand corner of the workspace card. <strong>Careful: this will delete all the saved objects in the workspace!</strong></p><p><strong>Copy data between workspaces</strong></p><p>Say you’ve got some sweet looking visualizations you want copied over to a new workspace. No problem! We got you covered. You’ll need to export all the objects from your old workspace and import them to your new workspace.</p><p><strong>Exporting Objects</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8ZWIJvbHWxpyIp6ii5sFgA.gif" /><figcaption>Exporting Objects</figcaption></figure><ol><li>Enter the “old” workspace, or the workspace that you would like to copy the objects from.</li><li>Click the Management Tab</li><li>Click the Saved Objects Option</li><li>Click the checkbox to select all objects, or selectively select the subset of objects you would like to migrate over</li><li>Click Export and Confirm</li><li>A .json file should be downloaded to you computer</li></ol><p><strong>Importing Objects</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*DDas-Vu7gxXLDaBy1d07nQ.gif" /><figcaption>Importing Objects</figcaption></figure><ol><li>Enter the “new” workspace, or the workspace that you would like to copy the objects into to.</li><li>Click the Management Tab</li><li>Click the Saved Objects Option</li><li>Click Import, and select the .json file generated in the export</li><li>Patiently wait a few moments!</li><li>Once complete, press Done</li></ol><p><strong><em>We hope you enjoy the new workspaces feature! Stay tuned for an exciting new feature we are working on to selectively pre-populate workspaces with even more “stock” dashboards including accreditation tables, station location analysis, station turnout dashboards, and more.</em></strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f6cc186a5ca5" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How Reliable Are You?]]></title>
            <link>https://medium.com/statengine/how-reliable-are-you-1a081ba7cd1c?source=rss-dfdc2f4de04f------2</link>
            <guid isPermaLink="false">https://medium.com/p/1a081ba7cd1c</guid>
            <category><![CDATA[fire-department]]></category>
            <category><![CDATA[analytics]]></category>
            <category><![CDATA[reliability]]></category>
            <category><![CDATA[programming]]></category>
            <dc:creator><![CDATA[Joe Chop]]></dc:creator>
            <pubDate>Wed, 01 May 2019 02:03:51 GMT</pubDate>
            <atom:updated>2019-05-01T03:21:03.632Z</atom:updated>
            <content:encoded><![CDATA[<p>In the fire service, <strong><em>reliability</em></strong> may be the most challenging, but also most valuable and insightful metric to collect. Reliability can be defined as “<strong>the probability that the required number of competently prepared staff and properly equipped apparatus will be available when a fire or emergency call is received (1)</strong>”. To simplify, when can say — when a call occurs in Station 11’s jurisdiction, how often does the home engine, Engine 11, respond to the call?</p><p>As the number of incidents increases, a station becomes less reliable as more of the stations’ apparatus are committed to these calls. Tracking reliability, and the corresponding performance metrics, including response time, across multiple facets, such as hour of the day, and call type leads to deeper insights and questions such as:</p><ul><li>When a station is committed, how is our response time impacted?</li><li>When, and at what hours, should we place additional resources including personnel or apparatus?</li><li>Where should the next station be built?</li></ul><p>In NFORS, we can not only track reliability, completely customizable to your departments organization, but we can also quantify the impact it has on our performance with real historic data.</p><p>Lets get started!</p><p><strong>For today’s exercise, we are going to calculate station reliability, however the same principals and guidance can be applied to build similar tables for apparatus level reliability.</strong></p><p><strong>Incident Count and Response Time By First Due</strong></p><ol><li>Create a new <strong><em>Enhanced Table </em></strong><em>v</em>isualization, using your incident index</li><li>Adjust the top time range to a longer time range, such as<strong> 2 years</strong></li><li>Click <strong>Add metrics, </strong>and click <strong>Metric</strong></li><li>Type/select <strong>Percentile</strong> as the Aggregation</li><li>Type/select <strong>durations.response_time.minutes</strong> as the field</li><li>Change one of the percents to <strong>90</strong>, and remove all other percents</li><li>Add a label of <strong>Response Time</strong></li><li>Under <strong>Buckets</strong>, select <strong>Split Table</strong></li><li>Select <strong>Terms</strong> as the Aggregation</li><li>Type/select <strong>address.first_due</strong> as the field</li><li>Change the size to at least the number of stations you have in your department, such as <strong>100</strong></li><li>Add a label of <strong>First Due Station</strong></li><li>Press play</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*MXq-W-MjIzmWU09OVbMVYw.gif" /><figcaption>Incident Counts and Response Times By First Due</figcaption></figure><p>You should now have multiple tables for each station.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*B9p3wRSDjZrWA78U_3rpyQ.png" /><figcaption>Example Table</figcaption></figure><p>Each table represents how many calls occurred in each station’s jurisdiction, and the corresponding 90th percentile response time.</p><p><strong>Filtering By Station Units</strong></p><ol><li>Click <strong>Add sub-buckets</strong></li><li>Select <strong>Split Rows</strong></li><li>Select <strong>Filters</strong> as the Sub Aggregation</li><li>For Filter 1, define a custom filter for a station. For example, if Station 11 is staffed primarily with E11, we may want to add a filter such as ‘<strong>description.units:E11</strong>’</li><li>Add a label, such as <strong>Station 11 Primary Unit</strong>s</li><li>Click <strong>Add Filter</strong></li><li>For Filter 2, define another station filter. For example, if Station 10 is staffed with two primary units, T10 and R2, we can add a filter such as ‘<strong>description.units:T10 OR description.units:R2</strong>’</li><li>Press Play</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*pbcVtbRGka9kox8ODKVEEg.gif" /><figcaption>Adding Station Filters</figcaption></figure><p>Your tables should now be updated with multiple rows, one row for each station filter.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zjPpihCiDBrhWlI6rue_cQ.png" /><figcaption>Example Table</figcaption></figure><p>As an example, the above table shows that of all the calls in Station 11’s jurisdiction:</p><ul><li>Units from Station 11 (i.e. E11), were dispatched on 5440 of the calls, and the 90th percentile response time was 6.4 minutes</li><li>Units from Station 10 (i.e. T10 or R2), were dispatched on 27 of the calls, and the 90th percentile response time was 10.5 minutes</li></ul><p>These tables only have rows/filters for two stations, Station 11 and Station 10, but the next steps would be to continue building out the custom filter logic to your departments specific organization. Building these filters can be burdensome at first, so we encourage to save often!</p><p><strong>Pro Tip — Calculating Percentages</strong></p><p>Once the rest of the station filters have been defined, the next step is to view reliability as percentages. This can can be a bit tricky, so follow carefully:</p><ol><li>Click<strong> Add Sub-buckets</strong></li><li>Select <strong>Split Rows</strong></li><li>Select <strong>Filters</strong> as the Sub Aggregation</li><li>Type * as Filter 1</li><li>Hover over the double arrow symbol, dragging the filter to be the second aggregation in the chain</li><li>Click the <strong>Options</strong> tab,</li><li>Under hidden columns, type <strong>0</strong></li><li>Click <strong>Show metrics for every bucket/level</strong></li><li>Click Play</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*g12chqVlHDn56ByxLzUpuQ.gif" /><figcaption>Adding Wildcard Filter</figcaption></figure><p>We just added a wildcard filter, which adds a wildcard, or unfiltered, view to the rows.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*T9oqBcByPi1fTibcrDoEuA.png" /><figcaption>Example Table</figcaption></figure><p>Continuing our example, the above table now has three additional columns</p><ul><li><strong>New</strong>: Column 1 (col1) is a label of the unfiltered view</li><li><strong>New</strong>: Column 2 (col2) shows the total number of calls for the entire station</li><li><strong>New</strong>: Column 3 (col3) shows the response time for the the entire station</li><li>Column 4 (col4) is a label of the filtered view (varies by row)</li><li>Column 5 (col5) shows the the total number of calls for the first due station, in which units from a particular station responded (varies by row)</li><li>Column 6 (col6) shows the the response time for the first due station, in which units from a particular station responded (varies by row)</li></ul><p>With the unfiltered view in context, we can now perform row-based computation. Lets add two simple calculations to show the reliability percentage, and response time difference.</p><ol><li>Click<strong> Add Computed Column</strong></li><li>Add a label of <strong>Reliability</strong></li><li>Type the formula, <strong>col5 / col2</strong></li><li>Change Pattern to <strong>0.0%</strong></li><li>Click <strong>Add Computed Column</strong></li><li>Add a label of <strong>Response Time Diff</strong></li><li>Type a formula, <strong>col3 — col6</strong></li><li>Change Pattern to <strong>0.0</strong></li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fpZn_Xsc6tYjQswoKqxCgw.gif" /><figcaption>Adding Computed Columns</figcaption></figure><p>With our sample data, the table now has the two new calculated columns:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*YRY3U4oBYXbmOm9OhAb-mA.png" /><figcaption>Example Table</figcaption></figure><p>Let’s review some quick insights:</p><p>For the 8,244 total calls (Column 2) in Station 11&#39;s first due:</p><ul><li>a <strong>primary unit from Station 11 responded 66%</strong> of the time (Column 7, Row 1)</li><li>a <strong>primary unit from Station 10 responded .3% of the time</strong> (Column 7, Row 2). Note: this doesn’t imply that Station 10 was the only responding station, but it does imply that Station 10 had to travel into Station 11’s jurisdiction (either as the primary unit or as aid)</li></ul><p>90 percent of the time, when an incident occurs in Station 11’s jurisdiction, the first unit responds in 7.4 minutes or less (Column 3)</p><ul><li>when a primary unit from Station 11 responds, the first unit on scene, arrives .9 minutes <strong>FASTER </strong>(Column 8, Row 1), in 6.4 minutes or less (Column 6, Row 1)</li><li>when a primary unit from Station 10 responds, the first unit on scene, arrives 3.1 minutes <strong>SLOWER </strong>(Column 8, Row 2)<strong>, </strong>in 10.5 minutes or less (Column 6, Row 2)</li></ul><p>Just from this one exercise, we can quickly see performance insights such as above. I challenge you to create your own reliability tables, and take it the next step:</p><p>How does your reliability vary across day of week, or hour of day?</p><p>How does your reliability vary across incident type?</p><p>How does your reliability vary across fire box, or response zone?</p><p><strong><em>Have questions? Feel free to reach out to us or share your story with the rest of the team!</em></strong></p><p>References<em>:</em></p><ol><li><em>FIRE SERVICE DEPLOYMENT: Accessing Community Vulnerability. Available at: </em><a href="https://www.nfpa.org/-/media/Files/Membership/member-sections/Metro-Chiefs/UrbanFireVulnerability.ashx?la=en"><em>https://www.nfpa.org/-/media/Files/Membership/member-sections/Metro-Chiefs/UrbanFireVulnerability.ashx?la=en</em></a></li></ol><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1a081ba7cd1c" width="1" height="1" alt=""><hr><p><a href="https://medium.com/statengine/how-reliable-are-you-1a081ba7cd1c">How Reliable Are You?</a> was originally published in <a href="https://medium.com/statengine">statengine</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Under the Weather]]></title>
            <link>https://medium.com/statengine/under-the-weather-41fe2f234885?source=rss-dfdc2f4de04f------2</link>
            <guid isPermaLink="false">https://medium.com/p/41fe2f234885</guid>
            <category><![CDATA[weather]]></category>
            <category><![CDATA[data-visualization]]></category>
            <category><![CDATA[fire-department]]></category>
            <dc:creator><![CDATA[Joe Chop]]></dc:creator>
            <pubDate>Sat, 30 Mar 2019 23:53:49 GMT</pubDate>
            <atom:updated>2019-04-01T03:18:26.470Z</atom:updated>
            <content:encoded><![CDATA[<p>Earthquakes, hurricanes, tornadoes, blizzards, ice storms, flooding, power outages, extreme heat, freezing cold, even volcanic eruptions, you name it — chances are your department has experienced a number of extreme weather events. Think about <strong>your</strong> last severe weather event. You probably remember — “Yeah, we were busy”. Or maybe….“it took longer to get there”.</p><p>Lets start exploring how we can use data to not only back these statements up, but start <strong>quantifying </strong>the impact of weather on performance.</p><p><strong>Background</strong></p><p>Most CAD or RMS systems do not natively capture weather data, so any weather analysis requires extensive data manipulation, curation, table joining, and purchasing weather data from third-party providers. Fortunately, <a href="https://www.nfors.org/">NFORS</a> takes away all these pains by automatically capturing and saving weather details for every incoming incident. The weather attributes collected include temperature, wind speed, wind gust, humidity, precipitation intensity, precipitation accumulation, and even textual summaries. Thanks to our third-party provider, <a href="https://darksky.net/">DarkSky</a>, we also populate weather on historic data, regardless of when you joined NFORS!</p><p><strong>Impact on Call Volume</strong></p><p>For the first visualization exercise, lets explore the relationship between extreme cold temperatures and call volume over time.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7EOVRAXu3gdcqX4vFaxOHQ.gif" /><figcaption>Extreme Temperatures and Call Volume</figcaption></figure><ol><li>Create a new <strong><em>Line</em></strong> <em>v</em>isualization, using your incidents index</li><li>Adjust top time range to a longer time range, such as<strong> 2 years</strong></li><li>Click <strong>Add metrics</strong></li><li>Select <strong>Y-Axis</strong></li><li>For aggregation, select <strong>Min, </strong>because we are visualizing minimum temperature</li><li>For field, type/select <strong><em>weather.currently.temperature, </em></strong>corresponding to<strong><em> </em></strong>the temperature at event time</li><li>Add a custom label of <strong>Min Temp</strong></li><li>Under Buckets, select <strong>Date Histogram</strong></li><li>Add a custom label of <strong>Time</strong></li><li>Change to the <strong>Metrics &amp; Axes </strong>tab</li><li>Expand the <strong>Min Temp</strong> section</li><li>On <strong>Value Axis</strong>, select a new axis<strong>, Right-Axis 1</strong></li><li>Press the play button</li></ol><p>Look at your results closely. Are there peaks or trends visible? For this department, we can see an extreme cold during the first week of 2018:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ufMfITvUZ0QSNAHqQOCHUA.png" /><figcaption>Peak in Extreme Temperature</figcaption></figure><p>And a corresponding peak in call volume:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zkGKMXHu4nBUWdIzjHlSLA.png" /><figcaption>Peak In Call Volume</figcaption></figure><p><strong>Going Further</strong></p><p>Extreme weather will likely affect a subset of call types more than the entire dataset. For example, such as in this case, cold weather will significantly affect the number of fire alarms, utility incidents (due to bursting pipes), and possibly car accidents due to treacherous road conditions. Other extreme events, such as heat waves, might not see the same extreme peaks in overall call volumes, but just an increase in a particular subset of EMS conditions such as heat strokes or dehydration. We recommend applying type, subtype, or category filters directly on visualizations or combining with existing dashboard visualizations to encourage deeper exploration.</p><p><strong>Other Extremes</strong></p><p>Temperature isn’t the only variable we may want to look at. How about max snow accumulation or wind speed?</p><p>For those of you in Florida, you may want to explore the impacts of hurricanes such as Hurricane Irma, which nearly doubled call volume for a week, in September, 2017.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*IqnX_4HvL6r3AYomxe0hPQ.png" /><figcaption>Wind Gust vs Call Volume <strong>(spike corresponds to Hurricane Irma)</strong></figcaption></figure><p><strong>Impact on Response Times</strong></p><p>When analyzing weather, do not be restricted to simply the impact on call volume. Impassable roads, reduced visibility, and simply driving at safer, slower speeds in bad weather conditions also affects response times. Here’s another simple visualization that compares the daily weather summary to response time.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*JZqSsx8xY90yF0v-oGz-gg.gif" /><figcaption>Weather Impact on Response Time</figcaption></figure><ol><li>Create a new <strong><em>Horizontal Bar </em></strong><em>v</em>isualization, using your incidents index</li><li>Adjust top time range to a longer time range, such as<strong> 2 years</strong></li><li>Change the <strong>Metric</strong> to <strong>Percentile</strong></li><li>Set the field to <strong><em>durations.travel.minutes</em></strong></li><li>Remove all other percentiles, and change <strong>99</strong> to <strong>90</strong></li><li>Add a custom label of <strong>Travel Time</strong></li><li>Under <strong>Buckets</strong>, select <strong>X-Axis</strong></li><li>Select <strong>Terms</strong></li><li>Type/select <strong><em>weather.currently.icon, </em></strong>which represents the weather category, or representative visual icon, at event time</li><li>Add a custom label of <strong>Weather</strong></li><li>Change <strong>Size</strong> to <strong>25</strong>, to ensure all weather categories are represented</li><li>Change to the <strong>Panel Settings</strong> tab, and click <strong>Order Buckets By Sum</strong></li><li>Press the play button</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*X8tvnA772ogUXBCjKqMdtw.png" /><figcaption>Travel Times in Different Weather Conditions</figcaption></figure><p>As expected, you may see slower travel times in adverse conditions such as snow, sleet, and rain. For this department, we see more than a 40% increase in travel times in sleet conditions when compared to the norm of cloudy/clear times.</p><p><em>The first step in preparing for the next weather event is understanding how the weather has impacted </em><strong><em>YOUR </em></strong><em>department in the past. Every department is impacted by weather differently, due to a number of controllable and uncontrollable factors including severity of the weather, geospatial features, and overall department readiness. The only way to quantify this impact, is to dive deep and start exploring your own data with exploration, visualizations, and detailed data analysis. We’re excited to hear your success stories, and we’re always hear to help! Reach out to us for questions, thoughts, or to simply showcase your own visualization and weather insights with the rest of the NFORS community.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=41fe2f234885" width="1" height="1" alt=""><hr><p><a href="https://medium.com/statengine/under-the-weather-41fe2f234885">Under the Weather</a> was originally published in <a href="https://medium.com/statengine">statengine</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Impact of Graveyard Shift]]></title>
            <link>https://medium.com/statengine/impact-of-graveyard-shift-65b01c7b4312?source=rss-dfdc2f4de04f------2</link>
            <guid isPermaLink="false">https://medium.com/p/65b01c7b4312</guid>
            <category><![CDATA[data-visualization]]></category>
            <category><![CDATA[fire-department]]></category>
            <category><![CDATA[data-science]]></category>
            <dc:creator><![CDATA[Joe Chop]]></dc:creator>
            <pubDate>Fri, 01 Mar 2019 02:52:04 GMT</pubDate>
            <atom:updated>2019-03-01T02:55:09.949Z</atom:updated>
            <content:encoded><![CDATA[<p>Anecdotally, we all think turnout times are increased during the graveyard hours. Crews are sleeping, need to dress, and have to travel from quarters to the responding apparatus, which might require traversing several flights of stairs or a fire pole. Several past studies [1, 2, 3] have confirmed our theory, some concluding that time of day has the greatest impact on response times, even more so than station design. In one study, across all station designs, the graveyard shift, defined as midnight to 6 AM, had increased turnout times between 50.8% and 58.9%.</p><p>Lets explore two methodologies on how to quantify the impact of graveyard runs within our own department.</p><h3>Filtering For Graveyard Hours</h3><p>A naive, but useful visualization is to simply calculate 90th percentile turnout for the two blocks of time</p><ul><li>Graveyard hours, midnight to 6 am</li><li>Non-graveyard hours, 6 am to midnight</li></ul><ol><li>Create a new <strong><em>Metric</em></strong> Visualization, using the incidents index</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*_FI0pJzD6G-qwifY9wRZJA.gif" /><figcaption>Create new metric visualization</figcaption></figure><p>2. First, visualize the overall 90th percentile turnout time</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*eSWyi1NCf7PMm59SzJIbFA.gif" /><figcaption>Calculating 90th Percentile Turnout</figcaption></figure><p>a) Increase time range to last 1 year</p><p>b) On <strong><em>Metrics</em></strong> panel, change count to <strong><em>Percentile</em></strong></p><p>c) For field, type <strong>durations.turnout.seconds</strong></p><p>d) Remove all other percents, and keep one at <strong><em>90</em></strong></p><p>e) Add a custom label, <strong><em>90th Percentile Turnout</em></strong></p><p>f) Press play</p><p>3. Now, lets break the metrics into two buckets, one for graveyard hours and the other for the non-graveyard hours</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-c_6z0yVpojF16SCaWyVcw.gif" /></figure><p>c) On <strong><em>Buckets</em></strong> panel, click split group, and choose a range <strong>Aggregation</strong></p><p>d) Change field to <strong><em>description.hour_of_day</em></strong></p><p>c) Create the two ranges</p><ul><li><strong>0–5</strong> (any incident opened between 0000 and 0559)</li><li><strong><em>6–24</em></strong> (any incident opened between 0600 and 2359)</li></ul><p>d) Add a custom label, <strong><em>Hour</em></strong></p><p>g) Press play</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5aOzKl4Qf-5GFRkhPr0EDQ.png" /><figcaption>Graveyard Turnout Times</figcaption></figure><p>The final visualization, now shows two distinct metrics. In this case, we show that 90% of the time, the first unit turnout is 114 seconds when the incident is opened between midnight and 6 AM. Alternatively, when the incident occurs during non-graveyard hours, the first unit turnout time is 85 seconds, 90% of the time.</p><h3><strong>Turnout Time by Hour of of Day</strong></h3><p>A more advanced, but another common methodology, often found in Standards of Cover is to illustrate turnout times and call volume, by hour of day in a line chart.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*JHg8x2vkQD2v9JH79h1YQA.png" /><figcaption>Example SOC</figcaption></figure><ol><li>Create a new <strong><em>Line</em></strong> Visualization, using the incidents index</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7haeYclSoVIleAhx-SNk4Q.gif" /><figcaption>Create new line visualization</figcaption></figure><p>2. First, visualize the count of incidents by hour of day</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*91kVdVdOs4G9wFB6NJIGbw.gif" /><figcaption>Incidents by Hour of Day</figcaption></figure><p>a) Increase time range to last 1 year</p><p>b) On Metrics panel, change the label to <strong><em>Incidents</em></strong></p><p>c) On Buckets panel, add X axis, select <strong><em>Terms</em></strong> aggregation, and add <strong><em>description.hour_of_day</em></strong></p><p>d) Change Order By to <strong><em>Alphabetical</em></strong></p><p>c) Change Order to <strong><em>Ascending</em></strong></p><p>e) Increase size to <strong><em>24</em></strong>, to account for 24 hours in a day</p><p>f) Add a custom label, <strong><em>Hour</em></strong></p><p>g) Press play</p><p>3. Now, lets add 90th percentile turnout as a second Y Axis</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ofOCEz3otRG5mhoqAVzxHA.gif" /><figcaption>Adding Turnout Times</figcaption></figure><p>a) On <strong><em>Metrics</em></strong> panel, click add <strong><em>Metrics</em></strong>, select <strong>Y Axis -&gt; Percentile</strong></p><p>b) For field, type <strong>durations.turnout.seconds</strong></p><p>c) Remove all other percentiles, and keep one at <strong><em>90</em></strong></p><p>d) Add a custom label, <strong><em>90th Percentile Turnout</em></strong></p><p>e) On Metrics &amp; Axes tab, expand the <strong><em>90th Percentile Turnout</em></strong> section, and change <strong><em>Value Axis</em></strong> to <strong><em>New Axis</em></strong></p><p>f) Press play</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*pa9bTZovykNRrUc7ufUEiA.png" /><figcaption>Turnout Time by Hour of Day</figcaption></figure><p>In either methodology, our suspicions are confirmed, from the hours of midnight to 6 AM, the 90th percentile turnout times increase from approximately 80 seconds to greater than 110 seconds, despite the decrease in call volume. In the second visualization, we gain additional insights — such as seeing the trend start earlier than midnight, closer to 10 PM.</p><p><em>How does your department turnout trend in the graveyard hours?</em></p><p><strong>References</strong></p><ol><li>Reglen, Dennis, and Daniel S. Scheller. “Improving Fire Department Turnout Times: Training v. Sanctions in a High Public Service Motivation Environment.”</li><li>Reglen, Dennis, and Daniel S. Scheller. “Fire department turnout times: A contextual analysis.” <em>Journal of Homeland Security and Emergency Management</em> 13.1 (2016): 167–189.</li><li>Upson, Robert, and Kathy A. Notarianni. <em>Quantitative evaluation of fire and EMS mobilization times</em>. Springer Science &amp; Business Media, 2012.</li></ol><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=65b01c7b4312" width="1" height="1" alt=""><hr><p><a href="https://medium.com/statengine/impact-of-graveyard-shift-65b01c7b4312">Impact of Graveyard Shift</a> was originally published in <a href="https://medium.com/statengine">statengine</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Visualizing Busy-ness]]></title>
            <link>https://medium.com/statengine/visualizing-busy-ness-928d9838e217?source=rss-dfdc2f4de04f------2</link>
            <guid isPermaLink="false">https://medium.com/p/928d9838e217</guid>
            <category><![CDATA[data-visualization]]></category>
            <dc:creator><![CDATA[Joe Chop]]></dc:creator>
            <pubDate>Wed, 30 Jan 2019 20:05:51 GMT</pubDate>
            <atom:updated>2019-01-30T20:09:00.435Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>Visualizing Busy-ness</strong></p><p>In a recently NFPA published <a href="https://www.nfpa.org/-/media/Files/Membership/member-sections/Metro-Chiefs/Urban-Fire-Forum/2018/2018UFFFireDepartmentResponse.ashx?la=e,%20nA">paper</a> analyzing fire department response metrics, the authors defined and discussed visualization techniques of a department’s workload or “busy-ness”. As a preliminary step, a fire department can explore this level of busy-ness by analyzing and visualizing the number of companies/units that are engaged on assignment per hour.</p><p>This sounds great in theory, but breaks down in practice. Attempting to build these types of visualizations natively with RMS, CAD, or even worse, manually using spreadsheet exports, will surely lead to headaches and frustration! Trying to do this dynamically, with varying time ranges, field filters, and streaming incoming data is downright impossible.</p><p>Fortunately, we recently integrated an exciting new Swimlane visualization. A <a href="https://github.com/prelert/kibana-swimlane-vis">Swimlane</a> visualization displays the behavior of a metric value over time aggregating across a data field. Each lane displays a different value of the selected field, with the relative size of the metric for that field over each time indicated by the color of the symbol at that time.</p><p>How might we do this in our Dashboards?</p><p><strong>Units Engaged Per Hour</strong></p><p>Lets visualize units engaged per hour as defined in the reference paper:</p><blockquote>“Units engaged per hour refers to the total number of emergency events to which apparatus are assigned within any one-hour time interval. Units engaged per hour is measured in one-hour intervals though the total time assigned for a particular unit may be shorter or longer than a single hour”</blockquote><ol><li>From the Visualization Tab, create a new Swimlane, and select the apparatus index</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*lL1ERBG3Mg2HDxgPtQzlnQ.gif" /><figcaption>Creating a new Swimlane visualization</figcaption></figure><p>2. Now set the visualization options</p><p>a) Set Metrics to <strong><em>Count</em></strong>, because we are “counting’ the number of responses over the time period in question.</p><p>b) Choose a <strong>Terms</strong> aggregation with field, <strong><em>apparatus_data.unit_id</em></strong> to display one unit per row. Don’t forget to increase the <strong><em>Size</em></strong> to the number of units you want to display</p><p>c) Add a sub aggregation on the time the event was opened, <strong><em>description.event_opened.</em></strong></p><p>d) Press the play button to preview. Don’t forget to increase the total time range on top menu bar if needed.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*QW95q56APOFJ-PjpS3tO0g.gif" /><figcaption>Visualization Options</figcaption></figure><p>You should now see a visualization similar to this:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*I7EEAjdEWd1OQ_BI7UGG3g.png" /><figcaption>Visualization Preview</figcaption></figure><p>4. Lets apply some styling tweaks</p><p>a) On options tab, select band thresholds (blue=1, yellow=2, red=3). A yellow rectangle indicates that there were 2 responses during the time interval.</p><p>b) Press play</p><p>c) Set the time interval as desired. For units engaged per hour, you would choose “hourly”, but for this particular department, we can instead look at a 3 hour interval, adapting for their lower call volume. For larger time ranges, such as months, you will want to choose a larger interval such as day.</p><p>You should now have a styled visualization similar to this:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*gPBPJiiGyp5-IiYUhpvNdQ.png" /><figcaption>Units Engaged Per 3 Hour</figcaption></figure><p>As you can tell, clear patterns and insights quickly emerge including</p><ol><li>E5 and E22 appear to be the busiest unit in terms of number of responses, generally with at least 1 response every three hours.</li><li>Strong vertical lines of reds and yellows down the chart, such as on Jan 23 07:00–10:00 indicate particular times of busy-ness. We can see E5, T3, E20, and several other units all had 3 responses during this 3 hour period.</li></ol><p><strong>Next steps….</strong></p><p>Of course, aggregating on <strong>Count</strong> and <strong>Unit Id</strong> is just one example of how to build Swimlane visualizations. We can increase the timeframe, for example to see if particular months are “busier” than other months, or even apply filters — to only look at overdose calls for example.</p><p>Unit engagement is just one example of a Swimlane, but the Swimlane visualization is also very useful for visualization trends over time such as:</p><ul><li><strong>Overall System Utilization (aggregated on a unit, first due station, or unit type)</strong></li><li><strong>Average or Max Turnout (aggregated on a unit, first due station, unit type, or incident type)</strong></li><li><strong>Travel Duration (aggregated on unit type, call priority, or incident type)</strong></li></ul><p><em>Have another use case for Swimlane visualizations?</em></p><p>— email us at <a href="/service/mailto:contact@statengine.io"><strong>contact@statengine.io</strong></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=928d9838e217" width="1" height="1" alt=""><hr><p><a href="https://medium.com/statengine/visualizing-busy-ness-928d9838e217">Visualizing Busy-ness</a> was originally published in <a href="https://medium.com/statengine">statengine</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Part 1: Accreditation Data Collection]]></title>
            <link>https://medium.com/statengine/part-1-accreditation-data-collection-afecd5f081ad?source=rss-dfdc2f4de04f------2</link>
            <guid isPermaLink="false">https://medium.com/p/afecd5f081ad</guid>
            <category><![CDATA[data-science]]></category>
            <dc:creator><![CDATA[Joe Chop]]></dc:creator>
            <pubDate>Tue, 30 Oct 2018 23:58:14 GMT</pubDate>
            <atom:updated>2019-06-27T11:52:16.077Z</atom:updated>
            <content:encoded><![CDATA[<p>Pulling data for accreditation purposes can be extremely difficult, confusing, time consuming, and expensive. Not any more!</p><p>One of our customers recently came to us with the goal of filling out this table specified by the <a href="https://cpse.org/wp-content/uploads/2018/03/CFAIITSpecs-10.01.16.pdf">CFAI Accreditation Model</a>. It probably seems daunting at first, but lets display this data in our Dashboard — all in less than 20 minutes.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1zsd1nt1GggK719Mqy1UQA.png" /></figure><p>So if we look at this table— we can isolate a few metrics that we need to collect:</p><ul><li>Alarm Handling</li><li>Turnout Time</li><li>Travel Time</li><li>Total Response Time</li><li>Effective Response Force Time</li><li>Effective Response Force Travel Time</li></ul><p><strong>Getting Started</strong></p><ol><li>Click the <strong>Visualize</strong> tab</li><li>Click the <strong>+</strong> button to create a new visualization</li><li>Select the <strong>Data Table</strong> option</li><li>Select the index pattern that contains *fire-incident*. You DO NOT want to select the index pattern that has the word apparatus in it.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3MVBnJfu5p5Rbr_Mq7rLMg.gif" /><figcaption>Creating a New Data Table</figcaption></figure><p><strong>Adding Metrics</strong></p><p>Now that we have a basic table lets report on our desired metrics.</p><p>Let’s add 90 percentile Alarm Handling</p><ol><li>Update the time range to a larger period of time (Last 5 years)</li><li>Click <strong>Add Metrics</strong>. Click the <strong>Metric</strong> Button</li><li>Select <strong>Percentiles</strong> as the Aggregation Type</li><li>Start typing “alarm handling” as the Field — select ‘<strong><em>durations.alarm_handling seconds</em></strong><em>’</em></li><li>Remove all other percentiles and modify one to <strong>90</strong>, which represents the 90th percentile</li><li>Add a custom label as “<strong>Alarm Handling</strong>”</li><li>Press play button</li><li>Save the visualization</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ui1Nl0sauyRPbx3aCgVkiw.gif" /><figcaption>Adding Metrics</figcaption></figure><p>Now lets repeat for the other metrics in question</p><ul><li>Turnout Time is represented as ‘<em>durations.turnout.seconds</em>’</li><li>Travel Time is represented as ‘<em>durations.travel.minutes</em>’</li><li>Total Response Time is represented as ‘<em>durations.total_response.minutes</em>’</li><li>Effective Response Time is represented as ‘<em>durations.effective_response_force_response.minutes</em>’</li><li>Effective Response Travel Time is represented as ‘<em>durations.effective_travel_force.minutes</em>’</li></ul><p>We should now have a table such as this</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*I9OC3xEQLTAggIul3IvcRw.png" /><figcaption>Basic Metrics</figcaption></figure><p><strong>Aggregating By Date</strong></p><p>Great, so we are now rolling up our desired metrics — but we now need to aggregate on several facets. Lets tackle the event date first by bucketing the data into Date ranges.</p><ol><li>Click <strong>Split Rows</strong></li><li>Select <strong>Date Histogram</strong> as Aggregation</li><li>Select <strong>Yearly</strong> as Interval</li><li>Add a custom label as “<strong>Year</strong>”</li><li>Press play button</li><li>Save the visualization</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1YSHBZnY2nG20Dxv_RFefg.gif" /><figcaption>Aggregating by Year</figcaption></figure><p><strong>Aggregating By Population Density</strong></p><p>Wow, getting somewhere! Lets now aggregate by Population Density.</p><ol><li>Click A<strong>dd sub-buckets</strong></li><li>Click <strong>Split-Rows</strong></li><li>Select <strong>Terms</strong> as Sub Aggregation</li><li>Start typing “population” as the Field — select ‘<strong><em>address.population_density</em></strong><em>’</em></li><li>Add a custom label as “<strong>Population Density</strong>”</li><li>Press play button</li><li>Save the visualization</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zF3ZgRLPOIRcqzCY4_Rvgw.gif" /><figcaption>Aggregating by Population Density</figcaption></figure><p><strong>Aggregating By Category/Response Class</strong></p><p>Looking good — but this table represents all incidents. We really need to aggregate across incident types or response class (e.g. Fire, EMS, etc).</p><ol><li>Click A<strong>dd sub-buckets</strong></li><li>Click <strong>Split-Table</strong></li><li>Select <strong>Terms</strong> as Sub Aggregation</li><li>Start typing “response_class” as the Field — select ‘<strong><em>description.response_class</em></strong><em>’</em></li><li>Add a custom label as “Response Class”</li><li>Press play button</li><li>Save the visualization</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*l3pNAZBGKVEApJl0F_z6LQ.gif" /><figcaption>Aggregating By Category</figcaption></figure><p><strong>Aggregating By Risk</strong></p><p>Finally, we will want to break the table by risk.</p><ol><li>Click A<strong>dd sub-buckets</strong></li><li>Click <strong>Split-Table</strong></li><li>Select <strong>Terms</strong> as Sub Aggregation</li><li>Start typing “risk_category” as the Field — select ‘<strong><em>description.risk_category</em></strong><em>’</em></li><li>Add a custom label as “Risk”</li><li>Press play button</li><li>Save the visualization</li></ol><p><strong>Wrapping Up</strong></p><p>Just like that — you have a starting point for your accreditation reporting. Stay tuned to Part 2 to see how we can filter out outliers, export to CSV, and more.</p><p><em>Need help with accreditation reporting?</em></p><p>— email us at <a href="/service/mailto:contact@statengine.io"><strong>contact@statengine.io</strong></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=afecd5f081ad" width="1" height="1" alt=""><hr><p><a href="https://medium.com/statengine/part-1-accreditation-data-collection-afecd5f081ad">Part 1: Accreditation Data Collection</a> was originally published in <a href="https://medium.com/statengine">statengine</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[New Dashboard Features]]></title>
            <link>https://medium.com/statengine/new-dashboard-features-4c4432255b43?source=rss-dfdc2f4de04f------2</link>
            <guid isPermaLink="false">https://medium.com/p/4c4432255b43</guid>
            <category><![CDATA[data-visualization]]></category>
            <dc:creator><![CDATA[Joe Chop]]></dc:creator>
            <pubDate>Tue, 30 Oct 2018 01:50:08 GMT</pubDate>
            <atom:updated>2018-10-30T21:25:32.412Z</atom:updated>
            <content:encoded><![CDATA[<p>We recently updated our dashboard component and we’re excited to share with you some of our new features. You may have noticed a slight change in color scheme and look and feel, but we’ve also added many new features including full-screen dashboard, filter autocomplete, multi-point polygonal geospatial filters, and the long awaited CSV Export!</p><p><strong>CSV Export</strong></p><p>The underlying data in any visualization in a dashboard can now be exported. This is useful for raw data analysis, import into Excel, outlier detection, and more.</p><ol><li>Click the three square symbol in the top right corner next to any visualization. You may need to hover for a second.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/82/1*maUHOoGCOF_YUFv3hXi3EA.png" /><figcaption>Look for this three square symbol</figcaption></figure><p>2. Click the Inspect Option</p><p>3. Click Download CSV (Formatted Option)</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fv1CCUxTryDLbWZXu5aCJQ.gif" /><figcaption>CSV Export</figcaption></figure><p><strong>Filter Autocomplete</strong></p><p>Filters are now easier than ever to build with autocomplete.</p><p>Lets build a new filter by incident type</p><ol><li>Click add a filter</li><li>Start typing ‘description.category’ — this should autocomplete</li><li>Select ‘is’ as operator</li><li>‘FIRE, EMS, and OTHER’ will autocomplete for values — select as desired</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7VPiwsRHq_lG3pbVD0abww.gif" /><figcaption>Filter Auto-complete</figcaption></figure><p><strong>Full-Screen Realtime Dashboards</strong></p><p>You can now make any dashboard full-screen. This works great when used in combination with Auto-refresh intervals, which automatically refreshes the data periodically without human interaction. Some of our customers use this in station dashboards and other unattended displays.</p><ol><li>If using Auto-Refresh, click the Auto-Refresh button on the top of any dashboard and select an interval</li><li>Click the Full screen button on the top of the dashboard</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*oMJ3fmWtgeEcRb5G9HWMwg.gif" /><figcaption>Full Screen Realtime Dashboard</figcaption></figure><p><strong>Polygon Geospatial Filters</strong></p><p>Previously, the dashboard could only filter down to rectangles. You can now apply geospatial filters of any polygon, which works great when filtering down to response zones, station boundaries, and other geospatial areas of interest that are not rectangles.</p><ol><li>Click the hexagon symbol on any map</li><li>Click once to start a point</li><li>Drag and click again for every point necessary to form the polygon</li><li>Click back on the starting point to close the polygon</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*HixGp1sDk3PIUd3w83X_Wg.gif" /></figure><p>We’re continuously improving, seeking feedback, and developing features requested by you!</p><p>Stay tuned to hear about our new accreditation tables, input controls, and more….</p><p><em>Have a new feature in mind?</em></p><p>— email us at <a href="/service/mailto:contact@statengine.io"><strong>contact@statengine.io</strong></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4c4432255b43" width="1" height="1" alt=""><hr><p><a href="https://medium.com/statengine/new-dashboard-features-4c4432255b43">New Dashboard Features</a> was originally published in <a href="https://medium.com/statengine">statengine</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Diving deeper with AVL]]></title>
            <link>https://medium.com/statengine/diving-deeper-with-avl-7287fc28a19a?source=rss-dfdc2f4de04f------2</link>
            <guid isPermaLink="false">https://medium.com/p/7287fc28a19a</guid>
            <category><![CDATA[data-visualization]]></category>
            <dc:creator><![CDATA[Joe Chop]]></dc:creator>
            <pubDate>Sat, 22 Sep 2018 00:54:05 GMT</pubDate>
            <atom:updated>2018-09-22T00:54:05.776Z</atom:updated>
            <content:encoded><![CDATA[<p>Automatic vehicle location (AVL) is a process to automatically track and transmit the location of moving vehicles. AVL data comes in many shapes and forms, but generally contains a few key fields: <strong>unit/car id, latitude &amp; longitude, bearing, speed, and timestamp. </strong>Whenever a vehicle is in motion, these measurements are periodically recorded, transmitted back to a central store, and saved in data files. One of our StatEngine partners recently supplied us with some test AVL data, which we started exploring. With AVL data in hand, lets preview some of the nifty features we can do in our dashboards.</p><p>Lets start with a simple question — <strong>what is the average speed of all our units?</strong></p><p>Utilizing a metric visualization, we can calculate average unit speed when the unit is enroute:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/664/1*unfrz7mfWxqja4iDN0M4tg.png" /><figcaption>Average Speed</figcaption></figure><p>Lets compare that to the average unit speed when the unit is available:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/520/1*IA8mRj_ClJ9BkR4yH_pyeg.png" /></figure><p><strong>How about where our units spend the most time?</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*t9FHGg01ri2lA-IUf58VlQ.png" /><figcaption>Most Utilized Routes</figcaption></figure><p>Taking this a step further…Lets create a heatmap based on speed, effectively visualizing where our units drive the fastest.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7Kgrf5ZGmR2sNdWm3HlZJg.png" /><figcaption>Vehicle Speed</figcaption></figure><p>Now, let’s determine the max speed for each unit</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*FJ3WE_EGS5W2fBKJ9H2B1w.png" /><figcaption>Max Speed by Unit</figcaption></figure><p>This is just a preview of whats to come with our new AVL features. More advanced capabilities are also in the works — such as rule-based alerts when units pass speed thresholds, route optimization, and near-time asset tracking.</p><p>We’re looking for partners that are willing to supply AVL data, preferably in near or realtime.</p><p><em>Got an AVL integration in mind?</em></p><p>— email us at <a href="/service/mailto:contact@statengine.io"><strong>contact@statengine.io</strong></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7287fc28a19a" width="1" height="1" alt=""><hr><p><a href="https://medium.com/statengine/diving-deeper-with-avl-7287fc28a19a">Diving deeper with AVL</a> was originally published in <a href="https://medium.com/statengine">statengine</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>