91看片

 

鈥淓ven in this age of machine learning, humans ultimately still make the decisions 鈥� on the race track and in the board room. That means they must be able to interpret the data and follow up with action. It鈥檚 up to the user to decipher the data they are provided ..."鈥�

You mash the gas pedal and torque associated with rapid acceleration quickly pushes your body backward into the seat. Objects to your left and right become a blur. You鈥檙e at top speed, and frankly, it鈥檚 quite a rush.

A hairpin corner approaches. Decisions must be made. How hard should you brake? What鈥檚 the best angle to take the corner without compromising speed? A wrong decision here could spell disaster.

Drivers need to fully understand what their vehicle can and can鈥檛 do in order to make such decisions. Andrew Rains, who earned a marketing degree in 2015 from the 91看片, has developed a device 鈥� 鈥� that relays real-time data back to motorsports enthusiasts to help them not only understand what they are capable of but learn from this information in order to make better decisions down the road.

鈥淓ven in this age of machine learning, humans ultimately still make the decisions 鈥� on the race track and in the board room,鈥� said Rains, whose device sells for $450 and is the only product on the market that conveniently relays information from sensors via Bluetooth directly to a user鈥檚 cell phone. 鈥淭hat means they must be able to interpret the data and follow up with action. It鈥檚 up to the user to decipher the data they are provided, whether it鈥檚 an analysis of where he or she should have braked sooner, gotten back into the throttle later, or in the business world, explore where customer service can be better directed.

鈥淓veryone gets something different out of the data they receive, and ultimately, people see what they want to see in the data. That鈥檚 very difficult because of our natural vices in life and backgrounds to see what the data really is, or how it should be interpreted. And as a driving coach, that鈥檚 a challenging thing to teach.鈥�

How does it work? APEX Pro is a one-inch tall by four-inch long piece of hardware that mounts on the dash of the car and uses GPS and accelerometers inside the device to measure the accelerations/decelerations the car is experiencing. 鈥淚t inputs all of the data into a vehicle dynamics model, a physics model,鈥� Rains said. 鈥淲hat鈥檚 different about us is we use a nine-axis accelerator. Uphill, downhill, flat, proprietary, etc. That鈥檚 what tells you if you can brake deeper.鈥�

Rains was a racing enthusiast long before he envisioned APEX Pro, which won the $100,000 grand prize at the 2018 . The Birmingham native enjoyed a successful stint as a driver and team owner as a college student, competing in a number of SCCA-sanctioned endurance events from Barber Motorsports Park, Road Atlanta, to Laguna Seca. He even recorded a pair of podium (top three) finishes last year.

What do successful race car drivers do as soon as they climb out of their cars? In many cases, they reach for a cold soft drink. You鈥檝e probably done that a time or two this week as well. But soft drink bottles don鈥檛 just wind up in vending machines by accident, nor are they magically placed on your favorite grocer鈥檚 aisles. It鈥檚 a multi-faceted process, and a big part of that process involves recording real-time data. Sodas don鈥檛 need to brake at 90 mph or fully understand what apex to take the next twist in the road, but professionals in the industry track information provided by sensors in vending machines and on the manufacturing warehouse floor that help that can of soda be the best can of soda it can be.

The next time you stand in front of a soda vending machine, consider the amount of artificial intelligence inside. Other than simply popping out a Coke Zero, such machines have enough brainpower via sensors to detect issues like temperature drops, electricity failures, coin jams, bottle jams, low product, etc. This information is relayed back to the bottler, who can begin resolving the issue.

鈥淭he sensors inside these machines are just like the check engine light in your car,鈥� said Steve Mann, IT Systems Manager at , North Carolina, for the past 17 years who earned a degree in information systems management from the 91看片 in 1991.

鈥淭he types of sensors regulate temperature (38 degrees is desired), electricity, whether the lights are on or off outside the machine, and does product rotation inside the machine for us. As with any type of sensor, you can key the code and get back the information that tells you what鈥檚 wrong. Some machines can even call home and report that information based on what you want to monitor.鈥�

Whereas much data comes in the form of surveys or sales reports and allows management to make decisions for a company, sensors within machines in the manufacturing sector talk to one another. Data provided from one sensor provides information needed for another machine to perform a task. In the beverage industry, sensors insure that products are blended with the precise measure of ingredients and concentrate.

鈥淚f it鈥檚 a recipe and I鈥檓 blending a beverage into a specific sized container, I鈥檝e got to have this much concentrate, this much sugar, this much water, and so on, to mix at a certain temperature and then create your finished good.

鈥淥nce you release a wave of orders to the warehouse floor, sensors are not only picking up the UPC codes, but also all of the dimensions associated with it to make sure it can ship the product on the right order to assemble it on the correct pallet. Machine learning is so important here because all of those dimensions and information is used by the machines to make sure that you pack a stable pallet that can be delivered on a truck to a customer and then you can easily unload it and put it into a cooler or on the shelves.鈥�

Whereas machine learning nearly spelled the end of humanity in the 鈥淭erminator鈥� series, Mann reminded that professionals continue to monitor system performance, from the vending machines to all products on the warehouse floors.

鈥淭here are lots of dashboards and results that keep track of the quality that we are putting out,鈥� he said. 鈥淲e want to make as much as we can without having to waste anything. But machines help take the human error out. Data today is providing information that allows us to ask, 鈥楢re we running the right products at the right time?鈥� to 鈥楢re we producing it as quickly as we can?鈥� and 鈥業s there a better way to not only produce this product in Location A, but maybe we should be producing it in a different manufacturing facility, or maybe we should add a production line to help offset things.鈥� With all of the data run through systems today, there are plant managers and line specialists looking for better ways to make things more efficient. You have to be able to react to this data.鈥�

鈥淒ata today is providing information that allows us to ask, 鈥楢re we running the right products at the right time?鈥� to 鈥楢re we producing it as quickly as we can?鈥� ..."鈥�

Should data received, either via sensors or financial reports, always be adhered to?

鈥淲hen you ignore the data, just like in any business, and you鈥檙e making decisions based off of customer information, reports, or surveys, there has to be an element of intuition,鈥� said Rains. 鈥淐ustomers are saying this, the board wants this, and investors are telling you this. But if you鈥檙e in the seat and you have to make the decisions 鈥� then you have to aggregate that information and not just go with what you think or feel is best. That鈥檚 exactly how we work on the race track. We use the data as one element. But as a driving coach, I use it to support what I want to communicate to the drivers. The other half is 鈥� are they actually going to do it?鈥�

Though machine-learned data is vital to production on the manufacturing floor, Mann insists that data provided to board rooms is equally important to the non-machines (executives) running companies.

鈥淎 lot of executives get daily sales reports that tell them what was budgeted based on what pricing was and think of it more of a stock market of sales. It helps them to make better decisions."鈥�

鈥淎 lot of executives get daily sales reports that tell them what was budgeted based on what pricing was and think of it more of a stock market of sales. It helps them to make better decisions,鈥� Mann said with assurance. 鈥淒id we push the right products? Do we have the right price points in the field? Are we going after the market the right way? They can see what they forecasted, planned and priced versus what actually took place.

鈥淭hose reports have become critical for executives to make decisions and validate the decisions that they made. Business teams have come up with their own analytics teams because they are trying to analyze the data more and more than in the past. They can turn around and be more self-sufficient with data provided to them. Am I serving the customer the right way, not just selling the product? It鈥檚 responding to their issues, it鈥檚 responding to equipment repairs or services they need, complaints.鈥�

But don鈥檛 over-analyze.

鈥淚f you are trying to record every single decision you make or action you take, in business or in racing, and aggregate all of that data constantly 鈥� you鈥檙e going to end up with data overload and make poor decisions.鈥�