Let Trim do the hard work for you -- including negotiating your bills, contesting bank fees, canceling old subscriptions, getting you out of debt, and automatically building towards your savings goals

Trim

I'm going to show you a cool new feature in . .NET Core 3.0 Let's say you want to create a simple, lean executable you can build and drop on to a server. For an example we'll create a console app that opens a line of text, reads it and displays the output. First, let's create a new .NET Core app: dotnet new console This will scaffold a new console application. Now run it: dotnet run It should look something like this: I'm on a Mac here, but it doesn't matter as long as your development box has the Installed. .NET Core CLI This displays "Hello World" on the console. Now, lets create a file called file.txt: this is a file!
With some lines
whatever Doesn't matter what you put in here, as long as it has some text in it. Next we'll create a something that will read those lines and display them. Remove the "Hello World!" code and replace it with this: [] lines = System.IO. .ReadAllLines(@ );

        foreach ( in lines)
        { .WriteLine( + );
        } .WriteLine( );
System. .ReadKey(); string File "test.txt" string line Console "\t" line Console "Press any key to exit." Console This is pretty much your basic cookie cutter code for: opening up a file reading it into a string array loop through the array line by line print each line exit Pretty simple stuff. When I run it on my machine it looks like this: And that's great. But I'm on a Mac here, what if I want it to run on a Windows Machine? Linux? No problem, this is .NET Core right? We'll just publish it to multiple targets. But what if .NET Core isn't installed on the machine? What if I just want a simple executable I can run to read this file without a pile of files or .Net core installed? Publishing in .Net Core Lets back up a little. .NET Core has had publishing profiles for a long time. The idea behind "target" publishing is one of the biggest selling points of the platform. Build your app, then publish it for a specific target, Windows, OSX, or Linux. You can publish it a few different ways: - This means relies on a shared version of .NET Core that's installed on the Computer/Server. Framework Dependent Deployment - This doesn't rely on .Net Core being installed on the server. All components are included with the package (tons of files usually). Self Contained Deployment - This is very similar to a framework dependent deployment, but it creates executables that are platform specific, but require the .NET Core libraries. Framework Dependent Executables Ok, so what's this cool new feature I'm going to show? Well, when you do a self contained deployment it's cool because you don't need the runtime installed, but it ends up looking something like this: This is the application we just built published as a Self Contained Deployment for Windows. Yikes. Let's say you wanted to share this file reader application, and asking someone to copy all these files into a folder to run something to read a text file. It's silly. New Feature: Self Contained Executables So to build that self contained executable I ran the following command: dotnet publish -c release -r win10-x64 This should look pretty familiar to you if you've done it before. But .NET Core 3.0 has a cool new feature: dotnet publish -c Release / PublishSingleFile= p: true Using this flag, it will build something a little different: That's MUCH better. It's now a single exe and .pdb. If we look at the info, it's not super small: But it includes the .NET Core runtime with it as well. And here's another cool feature. dotnet publish -r win-x64 -c Release /p:PublishSingleFile= /p:PublishTrimmed= true true So in our example it doesn't change the size, but if you have a large complex application with a lot of libraries, if you just publish it to a single file, it can get HUGE. By adding the flag it will only extract the libraries you need to run the application. PublishTrimmed So when we copy the files to a Windows 10 machine, we have a nice small package: And we run it and it works! On a machine without .NET Core installed! and if I change my target: dotnet publish -r linux-x64 -c Release = = /p :PublishSingleFile true /p :PublishTrimmed true I can run it on a Linux server without .NET Core just as easily: Just remember on a Linux machine you won't need the .NET Core runtime, but you will need the installed. Prerequisites for .NET Core on Linux Conclusion So this is a cool feature of .NET Core 3.0. If you want to build trimmed down self contained executables for any of the platforms, you can do it easily with a couple of flags. This is great for those stupid simple things: console apps, data readers, microservices, or whatever you want to build and just drop on a machine with no hassle. I thought it was a cool feature to show. with questions or comments! Yell at me on Twitter

Creating and Trimming Self Contained Executables in .NET Core - A Detailed Guide

While a lot of time can be saved the way we run meetings, organizing meeting rooms is a good start. One of my team members asked me, “Why do I pay so much attention to small details in the meeting room?” I told him that it is equal to having one additional hire in the team. He smiled and asked, “How?” I asked him to choose any meeting room in the building. We both walked to one of the meeting rooms that he pointed. I showed him a clean board, and the set of items always kept ready, as shown in the picture below. I asked him to pick any marker and write. It wrote well. I started explaining to him that when I joined the company one year back and attended meetings every day, I observed the below: Boards were not clean, and dusters were missing. Many of the board markers were not writing. , notepads, pens, pencils, and other note-taking utils were missing. Post-it cables were either missing or not working. HDMI Remote team members were struggling to hear and see us. Members are not aware of the meeting end time. Math: (sum of time taken in hours to fix the above list * sum of total number of people in each meeting * total number of meetings per day * number of working days per year) * additional stress and distraction coefficient = the total number of unproductive hours per year. I could convince him that the above number of unproductive hours can be equal to a good hire. Our continuously improving practices: We have a contract with a company for maintenance. I created an excel checklist which has a list of refill and cleaning items. It will be signed every day by our maintenance staff. Any new observation get appended to the sheet. Remote team members and people working from home were struggling to hear and see us. We now have standup pods as shown in the image below. It has a for better viewing and a handy . Only the person with the mic can talk, which improved one conversation at a time. wide-angle webcam bluetooth speaker and mic Standup Pod I use a timer on my phone for most of the meetings that I attend. Sometimes after the first 30 minutes, I ask the group if we can extend it for another 10 minutes. To simulate the same, we are planning to keep big timers in each meeting room, which beeps after 30 minutes. Another idea is to restrict booking of the meeting rooms to 30 minutes only. After a meeting, we made a practice of switching off the lights. Apart from the energy savings, it is visually simple to see a free meeting room and pick one. While a lot of time can be saved the way we run meetings, the design and organization of meeting rooms is a good start.

Trim team noise — 1. Organize meeting rooms

Author:
(1) Mohammad Akhlaghi, Centro de Estudios de F´ısica del Cosmos de Aragon (CEFCA), Plaza San Juan 1, 44001, Teruel, Spain {mohammad@akhlaghi.org}. Table of Links Abstract and Intro Simulating the Exposure Map Acknowledgment and References Appendix 2. SIMULATING THE EXPOSURE MAP Within Gnuastro, the executable in charge of simulating the exposure map of a given list of pointings is called astscript-pointing- -simulate. As with all Gnuastro features, it has a complete documentation in Gnuastro’s manual[2] which is available in many formats (on the command-line, as web pages, PDF and etc). A complete and dedicated tutorial is also available within the tutorials chapter[3] of the manual. Three of the images produced in the tutorial (as of version 0.21) are displayed in Figure 1. The left image shows the exposure map that is produced from a five-point, “+”-shaped set of pointings on the sky, with one pointing in the center and four on the outer extents. The middle image uses one of the hooks that are available within the script to produce a noisy image and not an exposure map. This hook can also be used to insert simulated/real objects. The right image shows how another hook can be used to account for trimming of the outer edges (which sometimes do not get exposed enough due to vignetting). This hook can also be used to insert any random area of bad pixels or to simulate persistence in near infra-red detectors (these can be large). For its processing, this script needs two input files: 1) a table containing the RA and Dec of each pointing, 2) a FITS image from the camera that the dither pattern is meant for. The image has to contain the celestial world coordinate system (WCS, see Calabretta & Greisen 2002) headers. Having a WCS is very important, especially for cameras with a wide field of view, because distortions can become visible/significant. The tutorial and dedicated manual for this script are available in the Gnuastro manual and will always be up to date with the running version; therefore for the practical details, we encourage readers to follow the Gnuastro manual. The future changes in the script that break with this research note will be clearly designated in a “changes after publication” subsection of the section describing this script. This paper is available on arxiv under CC 4.0 DEED license. [2] https://www.gnu.org/software/gnuastro/manual [3] https://www.gnu.org/software/gnuastro/manual/html node/Tutorials.html Author: (1) Mohammad Akhlaghi, Centro de Estudios de F´ısica del Cosmos de Aragon (CEFCA), Plaza San Juan 1, 44001, Teruel, Spain {mohammad@akhlaghi.org}. Author: Author: (1) Mohammad Akhlaghi, Centro de Estudios de F´ısica del Cosmos de Aragon (CEFCA), Plaza San Juan 1, 44001, Teruel, Spain {mohammad@akhlaghi.org}. Table of Links Abstract and Intro Abstract and Intro Simulating the Exposure Map Simulating the Exposure Map Acknowledgment and References Acknowledgment and References Appendix Appendix 2. SIMULATING THE EXPOSURE MAP Within Gnuastro, the executable in charge of simulating the exposure map of a given list of pointings is called astscript-pointing- -simulate. As with all Gnuastro features, it has a complete documentation in Gnuastro’s manual[2] which is available in many formats (on the command-line, as web pages, PDF and etc). A complete and dedicated tutorial is also available within the tutorials chapter[3] of the manual. Three of the images produced in the tutorial (as of version 0.21) are displayed in Figure 1. The left image shows the exposure map that is produced from a five-point, “+”-shaped set of pointings on the sky, with one pointing in the center and four on the outer extents. The middle image uses one of the hooks that are available within the script to produce a noisy image and not an exposure map. This hook can also be used to insert simulated/real objects. The right image shows how another hook can be used to account for trimming of the outer edges (which sometimes do not get exposed enough due to vignetting). This hook can also be used to insert any random area of bad pixels or to simulate persistence in near infra-red detectors (these can be large). For its processing, this script needs two input files: 1) a table containing the RA and Dec of each pointing, 2) a FITS image from the camera that the dither pattern is meant for. The image has to contain the celestial world coordinate system (WCS, see Calabretta & Greisen 2002) headers. Having a WCS is very important, especially for cameras with a wide field of view, because distortions can become visible/significant. The tutorial and dedicated manual for this script are available in the Gnuastro manual and will always be up to date with the running version; therefore for the practical details, we encourage readers to follow the Gnuastro manual. The future changes in the script that break with this research note will be clearly designated in a “changes after publication” subsection of the section describing this script. This paper is available on arxiv under CC 4.0 DEED license. This paper is available on arxiv under CC 4.0 DEED license. available on arxiv [2] https://www.gnu.org/software/gnuastro/manual https://www.gnu.org/software/gnuastro/manual [3] https://www.gnu.org/software/gnuastro/manual/html node/Tutorials.html https://www.gnu.org/software/gnuastro/manual/html

Gnuastro: Simulating The Exposure Map of a Pointing Pattern - Simulating the Exposure Map

It's been a while since my last appearance, but I'm excited to be back and to share something truly special with you. In this article, we'll explore my top 10 Google Sheets features that are guaranteed to boost your productivity, speed up your workflow, and make your data handling more efficient. So, without further ado, let's dive into these game-changing tools! https://youtu.be/iHAvhGO0Df0?si=j6ePVf_Y5814jupV&embedable=true Autofill One of the simplest yet most effective features in Google Sheets is Autofill. Imagine you need to input numbers from 1 to 100—instead of typing each one, simply start the sequence, grab the blue dot at the corner of the cell, and drag down. Google Sheets will fill the rest for you. This works not just with numbers, but with days of the week and months of the year too, saving you a significant amount of time. Removing Whitespace and Duplicates When working with large datasets, cleaning up data is crucial. Google Sheets makes this easy with tools to eliminate unwanted whitespace and duplicates. Just select your range, use the "Trim whitespace" option under Data Cleanup, and voilà—clean, tidy cells. Similarly, the "Remove duplicates" feature ensures that only unique values remain, streamlining your data for better analysis. Alternating Colors Enhancing readability can significantly improve the way you interact with your data. The "Alternating colors" feature applies different colors to alternating rows, making your spreadsheet easier to read and reducing errors when reviewing large blocks of data. Embedding Sheets in Web Pages For those who wish to integrate Google Sheets with their websites, the ability to embed sheets is invaluable. Simply choose to embed your document, copy the provided HTML code, and paste it into your website. This allows you to display live, interactive data directly on your web pages. Data Validation for Dates Data integrity can be maintained by restricting input to certain types of data. For example, you can set a cell to only accept dates. This is done through data validation, where you select "Date" as the criteria, and Google Sheets will provide a handy date picker to ensure accuracy. Importing HTML Need to pull data from a webpage into your spreadsheet? The "ImportHTML" function is your friend. Just write the formula, specify the URL, and choose the table you wish to import. This feature not only saves time but also ensures you can quickly update your spreadsheet with the latest web data. Extracting Unique Values To identify unique entries from a range, the "Unique" function is perfect. It extracts distinct values from your selected range, allowing you to see just the unique information without any repeats. Quick Sheet Creation If you need to create a new spreadsheet quickly, just type "sheets.new" in your Chrome address bar. This shortcut instantly opens a new Google Sheet, helping you get to work right away without navigating through menus. Import Range Sometimes, you might need to share data from a spreadsheet without giving access to the entire file. "Import Range" allows you to import specific ranges from one Google Sheet to another, keeping the data synchronized without manual intervention. Google Translate in Sheets Google Sheets can even translate text directly within cells using the "Google Translate" function. Simply specify the cell, the source language, and the target language, and the translation updates in real-time. Bonus Tip Column Stats For data analysts, the "Column Stats" add-on provides quick statistical insights into data columns without needing to apply complex formulas. This tool is excellent for a high-level overview of data characteristics. Inserting Images in Cells A lesser-known but handy feature is the ability to insert images directly into cells. This functionality allows for a more organized display of visual data, which can be particularly useful for inventory lists, directories, and catalogs. I hope you find these tips helpful for making the most out of Google Sheets. If you enjoyed this article, please consider liking and sharing it. Your feedback is always welcome, so feel free to drop a comment below. Don't forget to subscribe for more useful content, and I look forward to seeing you in future articles! Thank you for your support and happy sheeting! It's been a while since my last appearance, but I'm excited to be back and to share something truly special with you. In this article, we'll explore my top 10 Google Sheets features that are guaranteed to boost your productivity, speed up your workflow, and make your data handling more efficient. So, without further ado, let's dive into these game-changing tools! https://youtu.be/iHAvhGO0Df0?si=j6ePVf_Y5814jupV&embedable=true https://youtu.be/iHAvhGO0Df0?si=j6ePVf_Y5814jupV&embedable=true Autofill One of the simplest yet most effective features in Google Sheets is Autofill. Imagine you need to input numbers from 1 to 100—instead of typing each one, simply start the sequence, grab the blue dot at the corner of the cell, and drag down. Google Sheets will fill the rest for you. This works not just with numbers, but with days of the week and months of the year too, saving you a significant amount of time. Removing Whitespace and Duplicates When working with large datasets, cleaning up data is crucial. Google Sheets makes this easy with tools to eliminate unwanted whitespace and duplicates. Just select your range, use the "Trim whitespace" option under Data Cleanup, and voilà—clean, tidy cells. Similarly, the "Remove duplicates" feature ensures that only unique values remain, streamlining your data for better analysis. Alternating Colors Enhancing readability can significantly improve the way you interact with your data. The "Alternating colors" feature applies different colors to alternating rows, making your spreadsheet easier to read and reducing errors when reviewing large blocks of data. Embedding Sheets in Web Pages For those who wish to integrate Google Sheets with their websites, the ability to embed sheets is invaluable. Simply choose to embed your document, copy the provided HTML code, and paste it into your website. This allows you to display live, interactive data directly on your web pages. Data Validation for Dates Data integrity can be maintained by restricting input to certain types of data. For example, you can set a cell to only accept dates. This is done through data validation, where you select "Date" as the criteria, and Google Sheets will provide a handy date picker to ensure accuracy. Importing HTML Need to pull data from a webpage into your spreadsheet? The "ImportHTML" function is your friend. Just write the formula, specify the URL, and choose the table you wish to import. This feature not only saves time but also ensures you can quickly update your spreadsheet with the latest web data. Extracting Unique Values To identify unique entries from a range, the "Unique" function is perfect. It extracts distinct values from your selected range, allowing you to see just the unique information without any repeats. Quick Sheet Creation If you need to create a new spreadsheet quickly, just type "sheets.new" in your Chrome address bar. This shortcut instantly opens a new Google Sheet, helping you get to work right away without navigating through menus. Import Range Sometimes, you might need to share data from a spreadsheet without giving access to the entire file. "Import Range" allows you to import specific ranges from one Google Sheet to another, keeping the data synchronized without manual intervention. Google Translate in Sheets Google Sheets can even translate text directly within cells using the "Google Translate" function. Simply specify the cell, the source language, and the target language, and the translation updates in real-time. Bonus Tip Column Stats For data analysts, the "Column Stats" add-on provides quick statistical insights into data columns without needing to apply complex formulas. This tool is excellent for a high-level overview of data characteristics. Inserting Images in Cells A lesser-known but handy feature is the ability to insert images directly into cells. This functionality allows for a more organized display of visual data, which can be particularly useful for inventory lists, directories, and catalogs. I hope you find these tips helpful for making the most out of Google Sheets. If you enjoyed this article, please consider liking and sharing it. Your feedback is always welcome, so feel free to drop a comment below. Don't forget to subscribe for more useful content, and I look forward to seeing you in future articles! Thank you for your support and happy sheeting!

Top 10 Favorite Google Sheets Features for Everyday Work

Hey there, Hackers! Today, we’re talking about the almighty backlink. Although it could sometimes seem like an afterthought, a good set of high-quality backlinks could take a good piece to the next level. If you can backlink like a pro, then all your writing will be more authoritative, look stronger, read more professionally, and, best of all, can help you reach the coveted front page of Google. So, how can you master backlinks? Let's break down a few tips and learn from some of our backlink black belts at HackerNoon. Are you ready to step into the Backlink Dojo? Use our template with backlink tips and more to get started. AP-Style Reporting 1. Keywords Matter (Make It Make Sense!) When backlinking, always remember that you should include any relevant keywords from the website you are linking to in the writing of your HackerNoon article and make that your point of contact. For example, if you’re linking to a source reference about Game of Thrones, make sure that “Game of Thrones” is what gets highlighted and linked in your HackerNoon article. This ensures that your audience knows what they’re getting linked to. In short, make sure your anchor text is reflective of what you’re sourcing. Take a look at ’s new piece about visualizing network graphs as Hamilton characters and how all their backlinks make semantic sense to their reference source about the play and various other media. @iswaryam https://hackernoon.com/analysis-of-network-graphs-visualizing-hamilton-characters-as-a-social-network?embedable=true 2. Backlink Quality Good writers present strong sources to back up their claims, and you should make sure you’re not just taking info from anywhere. If you find yourself writing about news, academic research, or anything that is highly evidence-based or makes strong claims, always make sure you’re linking to authoritative, high-quality sources to reinforce your writing. Plus, the stronger the source, the more Google will like your writing! Check out ’s new piece about AI and extinct species, where he links back to strong sources like the National Library of Medicine, BBC Earth, and CBS News. @zacamos https://hackernoon.com/the-role-of-ai-in-resurrecting-extinct-species?embedable=true 3. Backlink Diversity While it could be tempting to list one place as your one-stop shop for everything you’re writing about, it can actually hinder your writing in the long run if you’re not careful. Linking to one place repeatedly can be considered backlink spamming by Google and our editors, and it could hurt your piece’s chances of becoming a HackerNoon Top Story or reaching search engines. Diversifying your knowledge pool and linking back to a wide variety of places indicates that you’ve researched your topic well and can present useful backlinks for those who would like to follow in your footsteps. For a good example of backlink diversity, read ’s deep dive into the world of PrisonTok, where he finds great interviews and quotes from former prisoners and legal specialists about the world of incarcerated influencers. @drewchapin https://hackernoon.com/the-rise-and-real-value-of-prisontok?embedable=true 4. Use a Lot of Backlinks (But Know When To Trim It Down!) Now you know how to use backlinks effectively; however, the next level is knowing to use them. No one likes to read a mega-highlighted article filled with so many links that it looks like a virtual minefield of redirects for every minor claim and occurrence. A truly skilled writer knows when it’s time to backlink for the benefit of the reader and when to let the writing confidently do its own thing. If you find yourself swamped with important sources and reference material, don’t be afraid to make a nice “References” section at the bottom of your piece to look like a real professional! when deftly links back to important sources like GitHub, Intel, and Nvidia as needed throughout this piece and then intelligently creates a “References” section at the bottom with a whole host of high-quality backlinks should readers need them. @venkat2811 https://hackernoon.com/a-digestible-high-level-overview-of-cpu-and-gpu-capabilities?embedable=true Are you ready to step into the Backlink Dojo? Use our template with backlink tips and more to get started. AP-Style Reporting Now That You’re A Pro, Check Out The #crypto-api Writing Contest by CoinGecko. If you’re an expert in all things crypto and with expertise regarding cryptocurrency data APIs, try your hand at the for your shot at winning from a Submissions are open now and run until July 18th, 2024. Get started with a or the official . #crypto-api Writing Contest $1,000 prize pool. new draft #crypto-api template Click for more info. here Wanna See Why Writers Love Publishing With HackerNoon? With mountains of testimonials from across the web about our great editorial team, wide content variety, and superb distribution engine, it’s no wonder thousands of writers chime into the site daily, and millions of readers keep coming back for more. Check out our new to see why readers trust us and why new and returning writers love submitting to us. HackerNoon Page That’s all for today, folks. See you in cyberspace!

Learn How To Backlink Like a Pro From The HackerNoon Editors

হ্যাকারনুন সম্পাদকদের থেকে কীভাবে একজন প্রো লাইক ব্যাকলিংক করবেন তা শিখুন

Erfahren Sie von den HackerNoon-Redakteuren, wie Sie Backlinks wie ein Profi erstellen

Aprenda a crear vínculos de retroceso como un profesional con los editores de HackerNoon

Apprenez à créer des backlinks comme un pro grâce aux éditeurs HackerNoon

HackerNoon संपादकों से जानें कि कैसे एक पेशेवर की तरह बैकलिंक बनाएं

HackerNoon 編集者からプロのようにバックリンクする方法を学ぶ

HackerNoon 편집자로부터 전문가처럼 백링크하는 방법 알아보기

Aprenda como fazer backlinks como um profissional com os editores do HackerNoon

Узнайте, как профессионально использовать обратные ссылки, от редакторов HackerNoon

HackerNoon Editörlerinden Bir Profesyonel Gibi Nasıl Backlink Vereceğinizi Öğrenin

Tìm hiểu cách liên kết ngược như một chuyên gia từ các biên tập viên HackerNoon

从 HackerNoon 编辑那里学习如何像专业人士一样建立反向链接

2023 was quite a ride for the trucking industry. We faced rising fuel prices, economic roller coasters, freight recessions, driver shortages, and a battle with inflation, which just wouldn't quit. It's been tough out there. As the founder of a startup with dispatch services in the U.S., I've got a front-row seat to the challenges facing our industry. My company aggregates over 100,000 loads daily, so I've seen firsthand how these trends are impacting carriers. My conclusion: To survive, a trucking company needs to know what’s coming, be flexible, and use modern tools. Here’s how to adapt to the trends that will shape the U.S. trucking industry. Use Data to Avoid Losses Everyone in the industry is feeling the squeeze. The trucking industry is facing its toughest challenge since 2008. Truckstop reported that in 2023, 77% of owner-operators had to book shorter routes due to rising costs, while 65% had booked fewer loads altogether. The American Transportation Research Institute ranks the economy as the biggest industry challenge, followed closely by truck parking availability and fuel prices. At the beginning of 2023, our inner data of aggregated loads predicted potential industry issues. I realized that margins would fall, and many operators would go bankrupt. That happened to Convoy, Yellow Corp, and other big players who were shutting down and laying off staff en masse. To be among the survivors, we took a risk to change our business model and focused on using AI tools. Being able to predict what’s coming helped my startup generate profit for three months last year. Now, we are once again close to breaking even. Carriers who don't use analytics will continue to run loads at a loss. With data analytics, truckers can identify the most efficient routes, predict load volumes, delivery times, market trends, seasonal fluctuations, and many other factors.  Previously, this information was only available to large companies. Now even small carriers can gain insights with tools like Truckerpath, DAT, or Cargofy (disclaimer: I'm biased, it's my startup). Switch to a Subscription Model It’s the era of the carriers who offer Logistics-as-a-Service. Looking to cut costs, e-commerce and other companies are seeking LaaS tools to manage their transportation services through cloud-based platforms. Instead of handling the logistics themselves, they prefer to outsource it to a third party. And this is where you can come into play. I saw firsthand how well it works. Launching a subscription-based model in 2023 allowed us to generate a more stable revenue stream. However, the benefits of LaaS extend beyond financial stability. By adopting a subscription model, carriers improve their efficiency by implementing transportation management systems (TMS), accounting software, and data analytics. They can better analyze financial data, gain insights about shippers’ debts, and manage contracts –– all in one place. Also, carriers can offer value-added services to shippers, such as real-time tracking and visibility. Start Using AI As Much as Possible AI tools increase truck drivers' revenue by an average of 17% per year, our internal analytics shows. By analyzing big data on supply-demand dynamics, seasonality, and freight availability, AI algorithms can guide truckers toward high-paying loads that minimize empty miles. Predictive matching helps carriers dismiss loads that lead to deadhead miles and wasted time. AI is also transforming the role of truck dispatchers. Instead of making numerous phone calls to find viable loads, they use AI and get pre-matched opportunities with calculated rates. With AI’s help, an individual dispatcher can now efficiently handle 2-3 times as many trucks. I don’t believe AI will fully replace the human interaction required for negotiations. But, it dramatically reduces manual work and empowers dispatchers to secure better loads in a fraction of the time. Trim Your Team (Sorry) The trucking industry has seen its fair share of ups and downs when it comes to employment. When times are good, companies hire thousands of workers, but when the market takes a turn, layoffs can be just as common. The  American Trucking Associations reported that, in 2022, a whopping 8.4 million people worked in trucking-related jobs across the economy. 3.54 million of those people were truck drivers. In 2023, many large companies such as Flexport, Uber Freight, Flock Freight, and Lipsey Logistics had to trim their teams. As the founder of a trucking startup, I also had to make the tough choice to downsize my team by 30%. This decision was not taken lightly, but it was essential to ensure the company's survival. Otherwise, my company would go bust, and I’d have to dismiss 100% of my staff. 2023 was quite a ride for the trucking industry. We faced rising fuel prices, economic roller coasters, freight recessions, driver shortages, and a battle with inflation, which just wouldn't quit. It's been tough out there. As the founder of a startup with dispatch services in the U.S., I've got a front-row seat to the challenges facing our industry. My company aggregates over 100,000 loads daily, so I've seen firsthand how these trends are impacting carriers. My conclusion: To survive, a trucking company needs to know what’s coming, be flexible, and use modern tools. Here’s how to adapt to the trends that will shape the U.S. trucking industry. Use Data to Avoid Losses Use Data to Avoid Losses Everyone in the industry is feeling the squeeze. The trucking industry is facing its toughest challenge since 2008. Truckstop reported that in 2023, 77% of owner-operators had to book shorter routes due to rising costs, while 65% had booked fewer loads altogether. The American Transportation Research Institute ranks the economy as the biggest industry challenge, followed closely by truck parking availability and fuel prices. 77% 77% 77% biggest industry challenge , biggest industry challenge biggest industry challenge At the beginning of 2023, our inner data of aggregated loads predicted potential industry issues. I realized that margins would fall, and many operators would go bankrupt. That happened to Convoy, Yellow Corp, and other big players who were shutting down and laying off staff en masse. To be among the survivors, we took a risk to change our business model and focused on using AI tools. Being able to predict what’s coming helped my startup generate profit for three months last year. Now, we are once again close to breaking even. Carriers who don't use analytics will continue to run loads at a loss. With data analytics, truckers can identify the most efficient routes, predict load volumes, delivery times, market trends, seasonal fluctuations, and many other factors.  Previously, this information was only available to large companies. Now even small carriers can gain insights with tools like Truckerpath, DAT, or Cargofy (disclaimer: I'm biased, it's my startup). Carriers who don't use analytics will continue to run loads at a loss. With data analytics, truckers can identify the most efficient routes, predict load volumes, delivery times, market trends, seasonal fluctuations, and many other factors.  Previously, this information was only available to large companies. Now even small carriers can gain insights with tools like Truckerpath, DAT, or Cargofy (disclaimer: I'm biased, it's my startup). Switch to a Subscription Model Switch to a Subscription Model It’s the era of the carriers who offer Logistics-as-a-Service. Looking to cut costs, e-commerce and other companies are seeking LaaS tools to manage their transportation services through cloud-based platforms. Instead of handling the logistics themselves, they prefer to outsource it to a third party. And this is where you can come into play. I saw firsthand how well it works. Launching a subscription-based model in 2023 allowed us to generate a more stable revenue stream. However, the benefits of LaaS extend beyond financial stability. By adopting a subscription model, carriers improve their efficiency by implementing transportation management systems (TMS), accounting software, and data analytics. They can better analyze financial data, gain insights about shippers’ debts, and manage contracts –– all in one place. Also, carriers can offer value-added services to shippers, such as real-time tracking and visibility. Start Using AI As Much as Possible Start Using AI As Much as Possible AI tools increase truck drivers' revenue by an average of 17% per year, our internal analytics shows. By analyzing big data on supply-demand dynamics, seasonality, and freight availability, AI algorithms can guide truckers toward high-paying loads that minimize empty miles. Predictive matching helps carriers dismiss loads that lead to deadhead miles and wasted time. AI is also transforming the role of truck dispatchers. Instead of making numerous phone calls to find viable loads, they use AI and get pre-matched opportunities with calculated rates. With AI’s help, an individual dispatcher can now efficiently handle 2-3 times as many trucks. I don’t believe AI will fully replace the human interaction required for negotiations. But, it dramatically reduces manual work and empowers dispatchers to secure better loads in a fraction of the time. Trim Your Team (Sorry) Trim Your Team (Sorry) The trucking industry has seen its fair share of ups and downs when it comes to employment. When times are good, companies hire thousands of workers, but when the market takes a turn, layoffs can be just as common. The American Trucking Associations reported that, in 2022, a whopping 8.4 million people worked in trucking-related jobs across the economy. 3.54 million of those people were truck drivers. American Trucking Associations American Trucking Associations American Trucking Associations In 2023, many large companies such as Flexport, Uber Freight, Flock Freight, and Lipsey Logistics had to trim their teams. As the founder of a trucking startup, I also had to make the tough choice to downsize my team by 30%. This decision was not taken lightly, but it was essential to ensure the company's survival. Otherwise, my company would go bust, and I’d have to dismiss 100% of my staff.

2024 Will Be Tough for the U.S. Trucking Industry: Here’s How Carriers Can Adapt

Starting my job hunt in a new country felt like stepping into a massive library with no idea where to find the book I needed. Coming right after the neighboring country's violent invasion of Ukraine in 2022, I arrived with high hopes but quickly encountered a wall. The job market here was a whole new world, and my usual approach just wasn't cutting it. Friends and settlement organizations suggested trying online job boards such as Job Bank Canada and Indeed. The concept of Job Application Automation with a Human Touch I wasn't aiming to spam companies with my resume. Instead, I sought to automate certain aspects of the process to save time and target suitable jobs for me in a more personalized manner. Moreover, it wasn’t solely about finding a job; it was also about learning how to use technology and AI, in particular. In sharing this story, I hope to inspire others to look at their challenges differently and maybe even find new ways to tackle them. Here's how I pieced it together, combining RSS feeds, server-side scripts, custom Chrome extensions, and the innovative use of AI to enrich data and personalize applications: Step 1: RSS for Job Listings Daily, a script on a Linux server I set up would fetch RSS feeds from Job Bank Canada, filtering opportunities based on my predefined keywords. It would store the data gathered into a cache file that another part of the system would work with later. This step appreciates the enduring value of RSS, allowing me and other old-school guys to stay updated with minimal effort. So a big special thank you to the services and platforms like Job Bank for maintaining it. Technologies used PHP
PHP Simple XML
Linux Bash Scripting
Crontabs Step 2: The Manual Touch: Reviewing Matches Despite the automation intention, I chose to review the matches manually, ensuring that only the most relevant opportunities were considered, at least at the beginning. This blend of automation and personal oversight was enough to maintain quality in the job hunt. I figured out there were too many vague jobs that didn't sound legit to me. Technically, the script reads the cached file and displays the number of items found for each keyword. At first, I was copying job posting details manually, but I gave up doing so for each listing. Even so, sometimes the same position can be seen twice or more if it contains different keywords, so… Technologies used PHP Step 3: Crafting a Custom Chrome Extension Screenshot of the extracted data (link) ..to expedite the review process and eliminate duplicates, I developed a simple parsing Chrome extension. The details it captured included the position name, salary range, contact info, and other essential data. I opted to automate this aspect in later versions of the system and either discard or repurpose the extension. Once I obtained the data, if I deemed the listing worthwhile, a click would save it directly to a Google Spreadsheet. Initially, I didn't want to set up a separate database for this project. Therefore, this convenient and widely used Google product suited my needs, allowing me to maintain simplicity The extension: Install the extension View Github source code Technologies used Google Spreadsheets API
Google Chrome’s extension
JavaScript Step 4: Enriching Data with Server-Side Scripts and AI I then enhanced my spreadsheet's listings using a server-side script to extract insights from employer websites, such as career page addresses, additional contact details, and social media links. More technical details were also gathered, like the server software and programming languages used. The major breakthrough came from integrating OpenAI's API, which summarized websites and classified industries, enriching each saved job listing with valuable context. To optimize OpenAI token usage, I trimmed the content to focus on the most relevant sections for analysis, avoiding excess from footers or irrelevant links. Technologies used PHP cURL
Prompt Engineering
OpenAI API
Google Spreadsheets API Step 5: Personalizing Cover Letters and resumes objectives To avoid the impersonal nature of mass applications, I used OpenAI's API selectively to tailor cover letters, inserting placeholders for the company name, industry, and the software they utilize. When you know the product or service the company produces, you understand the value you may provide to it. That was the only way to stand out among other applicants who used the semi-cold outreach approach. This strategy demonstrated a genuine interest in each employer even if I visited the website automatically in less than a second. Technologies used OpenAI API
PHP Step 6: Generating PDFs for Applications Aware of the limitations and risks associated with direct links in emails, I opted for a more conservative approach: converting resumes and cover letters into PDFs. Despite my preference for digital simplicity, the goal was to ensure my applications reached their destination with a reduced chance of being sidelined into spam folders or worse. The generated PDF files were cached on the server side and attached to my emails. Technologies used PHP
DomPDF
Ubuntu Linux
JavaScript Step 7: Enhancing Communication and Email Dispatch After the initial steps of data enrichment and personalization, it is now time for the critical phase — communication. Upon selecting a potential opportunity, a predefined email template is loaded into a text area, ready for any final, personal touches. The interface was similar to Gmail's one, where email addresses and subject lines were populated automatically. I used the position name, my name, and the company name in the subject. Sometimes I had to visit the website of a company of my interest again to ensure I don’t like or sound mistaken. Anyway, I reviewed and, if necessary, customized each message before proceeding. This part can be automated later too. At least I thought I would do so. To further avoid the stigma of automated spam, emails were dispatched at random intervals. This careful timing was designed to mimic human irregularity, reducing the likelihood of triggering spam filters and increasing the chance of email deliverability. Technologies used Gmail’s SMTP
CronJobs
Water.css
PHP Step 8: Refining Email opening Tracking Another gem of my job application toolkit was an email tracking system. I implemented a technique to outsmart tracker caching, subtly changing the tracking image's color scheme embedded in my email signature with each dispatch. This ensured each email open was uniquely recorded, triggering a script to log the activity directly into my Google Spreadsheet against the respective employer. This method provided precise insights into who engaged with my emails and how many times, significantly informing the effectiveness of my approach. While this tactic faced limitations with Gmail, it proved highly effective with other providers, particularly corporate ones. Technologies used cURL
Google Spreadsheets API
PHP Conclusion Sending each resume out of over 400 felt like giving away a great book to someone who's already received 400 others. The rare responses  I got were like small beams of light—rare but motivating. Adding a tracking system helped me see if my book even stood out in the pile. It was a journey of patience, perseverance, and a personal test showing that job platforms like Job Bank might not be the best for IT jobs. It seems employers prefer traditional methods like recruiters, LinkedIn, or employee referrals. Automating my job search was a huge learning curve. It was more than just speeding up applications; it was about making technology an ally in a personal quest. By blending RSS feeds, a Chrome extension, AI, and clever email strategies, I made my job hunt more efficient yet personal. The experience taught me valuable lessons about using technology wisely and staying true to myself. It's not about replacing the personal touch with technology but enhancing it, showing that with the right approach, we can push beyond what we thought was possible. Starting my job hunt in a new country felt like stepping into a massive library with no idea where to find the book I needed. Coming right after the neighboring country's violent invasion of Ukraine in 2022, I arrived with high hopes but quickly encountered a wall. The job market here was a whole new world, and my usual approach just wasn't cutting it. Friends and settlement organizations suggested trying online job boards such as Job Bank Canada and Indeed. The concept of Job Application Automation with a Human Touch I wasn't aiming to spam companies with my resume. Instead, I sought to automate certain aspects of the process to save time and target suitable jobs for me in a more personalized manner. Moreover, it wasn’t solely about finding a job; it was also about learning how to use technology and AI, in particular. In sharing this story, I hope to inspire others to look at their challenges differently and maybe even find new ways to tackle them. Here's how I pieced it together, combining RSS feeds, server-side scripts, custom Chrome extensions, and the innovative use of AI to enrich data and personalize applications: Step 1: RSS for Job Listings Daily, a script on a Linux server I set up would fetch RSS feeds from Job Bank Canada, filtering opportunities based on my predefined keywords. It would store the data gathered into a cache file that another part of the system would work with later. This step appreciates the enduring value of RSS, allowing me and other old-school guys to stay updated with minimal effort. So a big special thank you to the services and platforms like Job Bank for maintaining it. Technologies used PHP PHP Simple XML Linux Bash Scripting Crontabs PHP PHP Simple XML Linux Bash Scripting Crontabs Step 2: The Manual Touch: Reviewing Matches Despite the automation intention, I chose to review the matches manually, ensuring that only the most relevant opportunities were considered, at least at the beginning. This blend of automation and personal oversight was enough to maintain quality in the job hunt. I figured out there were too many vague jobs that didn't sound legit to me. Technically, the script reads the cached file and displays the number of items found for each keyword. At first, I was copying job posting details manually, but I gave up doing so for each listing. Even so, sometimes the same position can be seen twice or more if it contains different keywords, so… Technologies used PHP PHP Step 3: Crafting a Custom Chrome Extension Screenshot of the extracted data (link) Screenshot of the extracted data ( link ) link ..to expedite the review process and eliminate duplicates, I developed a simple parsing Chrome extension. The details it captured included the position name, salary range, contact info, and other essential data. I opted to automate this aspect in later versions of the system and either discard or repurpose the extension. Once I obtained the data, if I deemed the listing worthwhile, a click would save it directly to a Google Spreadsheet. Initially, I didn't want to set up a separate database for this project. Therefore, this convenient and widely used Google product suited my needs, allowing me to maintain simplicity The extension: Install the extension Install the extension View Github source code View Github source code Technologies used Google Spreadsheets API Google Chrome’s extension JavaScript Google Spreadsheets API Google Chrome’s extension JavaScript Step 4: Enriching Data with Server-Side Scripts and AI I then enhanced my spreadsheet's listings using a server-side script to extract insights from employer websites, such as career page addresses, additional contact details, and social media links. More technical details were also gathered, like the server software and programming languages used. The major breakthrough came from integrating OpenAI's API, which summarized websites and classified industries, enriching each saved job listing with valuable context. To optimize OpenAI token usage, I trimmed the content to focus on the most relevant sections for analysis, avoiding excess from footers or irrelevant links. Technologies used PHP cURL Prompt Engineering OpenAI API Google Spreadsheets API PHP cURL Prompt Engineering OpenAI API Google Spreadsheets API Step 5: Personalizing Cover Letters and resumes objectives To avoid the impersonal nature of mass applications, I used OpenAI's API selectively to tailor cover letters, inserting placeholders for the company name, industry, and the software they utilize. When you know the product or service the company produces, you understand the value you may provide to it. That was the only way to stand out among other applicants who used the semi-cold outreach approach. This strategy demonstrated a genuine interest in each employer even if I visited the website automatically in less than a second. Technologies used OpenAI API PHP OpenAI API PHP Step 6: Generating PDFs for Applications Aware of the limitations and risks associated with direct links in emails, I opted for a more conservative approach: converting resumes and cover letters into PDFs. Despite my preference for digital simplicity, the goal was to ensure my applications reached their destination with a reduced chance of being sidelined into spam folders or worse. The generated PDF files were cached on the server side and attached to my emails. Technologies used PHP DomPDF Ubuntu Linux JavaScript PHP DomPDF Ubuntu Linux JavaScript Step 7: Enhancing Communication and Email Dispatch After the initial steps of data enrichment and personalization, it is now time for the critical phase — communication. Upon selecting a potential opportunity, a predefined email template is loaded into a text area, ready for any final, personal touches. The interface was similar to Gmail's one, where email addresses and subject lines were populated automatically. I used the position name, my name, and the company name in the subject. Sometimes I had to visit the website of a company of my interest again to ensure I don’t like or sound mistaken. Anyway, I reviewed and, if necessary, customized each message before proceeding. This part can be automated later too. At least I thought I would do so. To further avoid the stigma of automated spam, emails were dispatched at random intervals. This careful timing was designed to mimic human irregularity, reducing the likelihood of triggering spam filters and increasing the chance of email deliverability. Technologies used Gmail’s SMTP CronJobs Water.css PHP Gmail’s SMTP CronJobs Water.css PHP Step 8: Refining Email opening Tracking Another gem of my job application toolkit was an email tracking system. I implemented a technique to outsmart tracker caching, subtly changing the tracking image's color scheme embedded in my email signature with each dispatch. This ensured each email open was uniquely recorded, triggering a script to log the activity directly into my Google Spreadsheet against the respective employer. This method provided precise insights into who engaged with my emails and how many times, significantly informing the effectiveness of my approach. While this tactic faced limitations with Gmail, it proved highly effective with other providers, particularly corporate ones. Technologies used cURL Google Spreadsheets API PHP cURL Google Spreadsheets API PHP Conclusion Sending each resume out of over 400 felt like giving away a great book to someone who's already received 400 others. The rare responses  I got were like small beams of light—rare but motivating. Adding a tracking system helped me see if my book even stood out in the pile. It was a journey of patience, perseverance, and a personal test showing that job platforms like Job Bank might not be the best for IT jobs. It seems employers prefer traditional methods like recruiters, LinkedIn, or employee referrals. Automating my job search was a huge learning curve. It was more than just speeding up applications; it was about making technology an ally in a personal quest. By blending RSS feeds, a Chrome extension, AI, and clever email strategies, I made my job hunt more efficient yet personal. The experience taught me valuable lessons about using technology wisely and staying true to myself. It's not about replacing the personal touch with technology but enhancing it, showing that with the right approach, we can push beyond what we thought was possible.

How I Tried to Automate Job Applications as a Software Engineer in Canada (and Failed)

Authors:
(1) Jo˜ao Almeida, CEGIST, Instituto Superior T´ecnico, Universidade de Lisboa Portugal;
(2) Jos´e Rui Figueira, CEGIST, Instituto Superior T´ecnico, Universidade de Lisboa Portugal;
(3) Alexandre P. Francisco, INESC-ID, Instituto Superior T´ecnico, Universidade de Lisboa Portugal;
(4) Daniel Santos, CEGIST, Instituto Superior T´ecnico, Universidade de Lisboa Portugal. 4 Algorithmic framework In this section, we explain the algorithmic framework of our meta-heuristic. In Subsection 4.1, we give a general overview. In Subsection 4.2, we introduce the instance decomposition and increment procedure. In Subsection 4.3, we present the heuristic for generating the initial timetables. In Subsection 4.4, we address the guided local search component. In Subsection 4.5, we address the adaptive large neighbourhood component. In Subsection 4.6, we address the variable neighbourhood search component. 4.1 Overview Our hybrid meta-heuristic uses mechanisms from adaptive large neighbourhood search (ALNS), guided local search (GLS), and variable neighbourhood search (VNS). Its generic representation is presented in Figure 1. Its inputs are the full set of curricula, L, and three stopping criteria, S1, S2, and S3. Additional parameters are also necessary for its functions and procedures, which are addressed in the following subsections. Its output is a feasible schedule, s∗ . Instead of solving the entire instance from the start, we decompose it by curricula and increment it iteratively until the set of curricula being solved, L′ , is equal to the entire set of curricula, L. The decomposition mechanism functions in the following way: a set of lectures belonging to a set of curricula is added to the problem. Then, we find a new feasible solution, s∗ , considering the current set of lectures, I′. When we have a feasible solution, we add the next set of curricula. Let us note that the assignments of each iteration can be changed if necessary in the following iterations. Our objective function, f(s ∗ ), represents the number of hard constraint violations in the current solution. Thus, the neighbourhoods are searched to find if there is a neighbour with fewer violations. The neighbourhood structures are explained in detail in Subsection 4.5. However, instead of using a regular objective function in this search, we introduce mechanisms for penalising certain features in the solution, which are useful for escaping local optima. Accordingly, after a certain number of iterations, the feature penalty vector, λ, is updated. The features and the update procedure are explained in detail in Subsection 4.4. However, sometimes this penalty is not enough to escape some local optima. To avoid those, we include a shaking phase, which occurs after a certain number of penalty recalculations without improvements. The number of local searches before updating the penalty vector is the stopping criterion S1. The number of penalty weight recalculations which is attempted before inducing a shaking phase is the stopping criterion S2. The duration of the shaking phase is the stopping criterion S3. The pseudo-code for this meta-heuristic is presented in Algorithm 1. 4.2 Instance decomposition and increment The procedure to increment the instance is straightforward. Curricula are added to the instance until a specific stopping criterion is met, or the entire set of curricula is included. We explore two different stopping criteria. One receives as input a fixed number of curricula to be added in each iteration, ρ. The other adds curricula until a certain number of hard constraint violations, f, is reached. Their respective pseudo-code algorithms are given in the Appendix. Algorithm 8 presents the fixed increments procedure and Algorithm 9 the violations-based increment procedure. Whenever a new curriculum, ℓ ′′, is added, its lectures, i ∈ aℓ ′′ , must be scheduled and added to the current solution, s∗ . The function which returns the schedule for each new curriculum is presented in the following subsection. Figure 2 provides a diagram of the procedure. We test two different sorting algorithms for deciding the order in which the curricula are added. The random sorting procedure chooses a random curriculum. The ordering procedure picks a degree, then, by ascending order of year, picks a class, and then adds the first and second trimesters, each corresponding to a different curriculum, ℓ, or timetable. This procedure is presented in Algorithm 2. 4.3 Initial solutions 4.4 Guided local search Considering each type of constraint violation as an objective, this methodology can be analysed as a multi-objective search technique. Effectively, what is happening is that out of the objective functions being regarded, the most challenging to improve are identified. Then, their weights, λi , are updated to help the search move towards non-dominated solutions that improve these objectives’ performance, even if the total sum of the effective violations increases. 4.5 Adaptive large neighbourhood search In ALNS, the local search is based on different neighbourhood structures. In each iteration, a neighbourhood structure, n, is selected based on a probability vector, p, and a local search is conducted. If a better solution is found, it replaces the bestknown solution. Let us note that we use an augmented objective function adapted from GLS procedures, f ′ (s). Then, based on the current solution characteristics and the previous successes of searches in each neighbourhood, the probability vector, p, is updated. The general algorithm framework is presented in Algorithm 5. To implement ALNS, it is necessary to define the set of neighbourhood structures, N, the procedure for updating the probability vector, p, and the stopping criterion. We established six main neighbourhood structures, which we named: worst slot, worst curriculum, worst professor, worst room type, different day, and precedence neighbourhoods. Moreover, each of them has two sub-types, named swap and non-swap. In each neighbourhood, a lecture is selected, and then, in the swap sub-type, its time slot is swapped with another lecture, and in the non-swap sub-type, it is randomly assigned to another time slot. The neighbourhoods follow a general framework that can be applied to each type of constraint violation. The first step is randomly selecting a lecture potentially causing violations. Then, either an adequate new time slot or an adequate swap lecture is randomly assigned. This adequacy is assessed based on the type of violation being caused. This general framework is presented in Figure 4. Specifically, each neighbourhood structure operates in the following manner: 4.6 Variable neighbourhood search In VNS, whenever local optima are identified, a shaking phase is used to move to different areas of the solution space. In our meta-heuristic, such a shaking phase is conducted if no improvement moves are found after a certain number of feature penalty updates are made. This shaking phase functions similarly to the ALNS algorithm, but every move is accepted instead of only those which introduce improvements. This phase is stopped after a number of iterations, S3. It is presented in Algorithm 7. This paper is available on arxiv under CC 4.0 license. Authors: (1) Jo˜ao Almeida, CEGIST, Instituto Superior T´ecnico, Universidade de Lisboa Portugal; (2) Jos´e Rui Figueira, CEGIST, Instituto Superior T´ecnico, Universidade de Lisboa Portugal; (3) Alexandre P. Francisco, INESC-ID, Instituto Superior T´ecnico, Universidade de Lisboa Portugal; (4) Daniel Santos, CEGIST, Instituto Superior T´ecnico, Universidade de Lisboa Portugal. Authors: (1) Jo˜ao Almeida, CEGIST, Instituto Superior T´ecnico, Universidade de Lisboa Portugal; (2) Jos´e Rui Figueira, CEGIST, Instituto Superior T´ecnico, Universidade de Lisboa Portugal; (3) Alexandre P. Francisco, INESC-ID, Instituto Superior T´ecnico, Universidade de Lisboa Portugal; (4) Daniel Santos, CEGIST, Instituto Superior T´ecnico, Universidade de Lisboa Portugal. 4 Algorithmic framework In this section, we explain the algorithmic framework of our meta-heuristic. In Subsection 4.1, we give a general overview. In Subsection 4.2, we introduce the instance decomposition and increment procedure. In Subsection 4.3, we present the heuristic for generating the initial timetables. In Subsection 4.4, we address the guided local search component. In Subsection 4.5, we address the adaptive large neighbourhood component. In Subsection 4.6, we address the variable neighbourhood search component. 4.1 Overview Our hybrid meta-heuristic uses mechanisms from adaptive large neighbourhood search (ALNS), guided local search (GLS), and variable neighbourhood search (VNS). Its generic representation is presented in Figure 1. Its inputs are the full set of curricula, L, and three stopping criteria, S1, S2, and S3. Additional parameters are also necessary for its functions and procedures, which are addressed in the following subsections. Its output is a feasible schedule, s∗ . Instead of solving the entire instance from the start, we decompose it by curricula and increment it iteratively until the set of curricula being solved, L′ , is equal to the entire set of curricula, L. The decomposition mechanism functions in the following way: a set of lectures belonging to a set of curricula is added to the problem. Then, we find a new feasible solution, s∗ , considering the current set of lectures, I′. When we have a feasible solution, we add the next set of curricula. Let us note that the assignments of each iteration can be changed if necessary in the following iterations. Our objective function, f(s ∗ ), represents the number of hard constraint violations in the current solution. Thus, the neighbourhoods are searched to find if there is a neighbour with fewer violations. The neighbourhood structures are explained in detail in Subsection 4.5. However, instead of using a regular objective function in this search, we introduce mechanisms for penalising certain features in the solution, which are useful for escaping local optima. Accordingly, after a certain number of iterations, the feature penalty vector, λ, is updated. The features and the update procedure are explained in detail in Subsection 4.4. However, sometimes this penalty is not enough to escape some local optima. To avoid those, we include a shaking phase, which occurs after a certain number of penalty recalculations without improvements. The number of local searches before updating the penalty vector is the stopping criterion S1. The number of penalty weight recalculations which is attempted before inducing a shaking phase is the stopping criterion S2. The duration of the shaking phase is the stopping criterion S3. The pseudo-code for this meta-heuristic is presented in Algorithm 1. 4.2 Instance decomposition and increment The procedure to increment the instance is straightforward. Curricula are added to the instance until a specific stopping criterion is met, or the entire set of curricula is included. We explore two different stopping criteria. One receives as input a fixed number of curricula to be added in each iteration, ρ. The other adds curricula until a certain number of hard constraint violations, f, is reached. Their respective pseudo-code algorithms are given in the Appendix. Algorithm 8 presents the fixed increments procedure and Algorithm 9 the violations-based increment procedure. Whenever a new curriculum, ℓ ′′, is added, its lectures, i ∈ aℓ ′′ , must be scheduled and added to the current solution, s∗ . The function which returns the schedule for each new curriculum is presented in the following subsection. Figure 2 provides a diagram of the procedure. We test two different sorting algorithms for deciding the order in which the curricula are added. The random sorting procedure chooses a random curriculum. The ordering procedure picks a degree, then, by ascending order of year, picks a class, and then adds the first and second trimesters, each corresponding to a different curriculum, ℓ, or timetable. This procedure is presented in Algorithm 2. 4.3 Initial solutions 4.4 Guided local search Considering each type of constraint violation as an objective, this methodology can be analysed as a multi-objective search technique. Effectively, what is happening is that out of the objective functions being regarded, the most challenging to improve are identified. Then, their weights, λi , are updated to help the search move towards non-dominated solutions that improve these objectives’ performance, even if the total sum of the effective violations increases. 4.5 Adaptive large neighbourhood search In ALNS, the local search is based on different neighbourhood structures. In each iteration, a neighbourhood structure, n, is selected based on a probability vector, p, and a local search is conducted. If a better solution is found, it replaces the bestknown solution. Let us note that we use an augmented objective function adapted from GLS procedures, f ′ (s). Then, based on the current solution characteristics and the previous successes of searches in each neighbourhood, the probability vector, p, is updated. The general algorithm framework is presented in Algorithm 5. To implement ALNS, it is necessary to define the set of neighbourhood structures, N, the procedure for updating the probability vector, p, and the stopping criterion. We established six main neighbourhood structures, which we named: worst slot, worst curriculum, worst professor, worst room type, different day, and precedence neighbourhoods. Moreover, each of them has two sub-types, named swap and non-swap. In each neighbourhood, a lecture is selected, and then, in the swap sub-type, its time slot is swapped with another lecture, and in the non-swap sub-type, it is randomly assigned to another time slot. The neighbourhoods follow a general framework that can be applied to each type of constraint violation. The first step is randomly selecting a lecture potentially causing violations. Then, either an adequate new time slot or an adequate swap lecture is randomly assigned. This adequacy is assessed based on the type of violation being caused. This general framework is presented in Figure 4. Specifically, each neighbourhood structure operates in the following manner: 4.6 Variable neighbourhood search In VNS, whenever local optima are identified, a shaking phase is used to move to different areas of the solution space. In our meta-heuristic, such a shaking phase is conducted if no improvement moves are found after a certain number of feature penalty updates are made. This shaking phase functions similarly to the ALNS algorithm, but every move is accepted instead of only those which introduce improvements. This phase is stopped after a number of iterations, S3. It is presented in Algorithm 7. This paper is available on arxiv under CC 4.0 license. This paper is available on arxiv under CC 4.0 license. available on arxiv

TRIM

asktrim.com

-- employees

Since n.d.

COMPANY RANKING

TRIM

EVERGREEN INDEX #10822

Representing this company? Click here to claim and customize this page!

Trim WIKI

Apply to get your Tech Company News Page live on HackerNoon.com today!

TRIM

asktrim.com

-- employees

Since n.d.

COMPANY RANKING

TRIM

EVERGREEN INDEX #10822

RELATED COMPANIESRANK

STORIES ABOUT TRIM

Representing this company? Click here to claim and customize this page!

Trim WIKI

Apply to get your Tech Company News Page live on HackerNoon.com today!