powered by Consignor

The AI hype: How artificial intelligence improves e-commerce, logistics and transport

Posted: 13. June 2018-Likes: 0-Comments: 0-Categories: Ikke kategoriseret

The AI hype: How artificial intelligence improves e-commerce, logistics and transport

 

Personalized buying experiences, chat-bots, anomaly detection, image recognition and route planning. AI expert Lars Ritland guides us through how artificial intelligence is used in the industry, what benefits AI implies and how to implement it in your business.

Technology is constantly developing and getting better. Along with this, consumers are getting more demanding. Thus companies need to have efficient and innovative solutions in order to stay competitive, and that is where AI enters the picture.

“Artificial intelligence can automate, quality assure and optimize processes, analyze huge amounts of data, solve previously unsolvable problems or come up with new and value creating features for companies and customers. AI augments human capabilities but also eliminates routine work, which allow human workforces to do more meaningful and value creating work,” Lars Ritland, AI expert at Consignor, says.

At a more practical level, Lars lists examples of how AI can be used in:

Lars Ritland

Lars Ritland is implementing AI at software company Consignor.

  • Ecommerce: the recommendation service in online shops that show you recommended products based on the content of your shopping basket. Customer service chat-bots. Churn prediction, which detects potentially leaving customers.
  • Logistics: Image recognition, AI can make it possible to extract information from a picture i.e. a parcel’s size (and weight). Anomaly detection. Quality assuring through monitoring of pictures, video and data.
  • Transport/delivery: Route planning based on weather forecasts and traffic conditions, prediction of delivery times, fraud detection cameras, delivery robots.
Read more: Tesla’s Semi Truck could revolutionise the transport industry

 

Along with the AI hype, many companies want to be the first to go to market with AI capabilities and in their eagerness reposition many of their existing conventional solutions as AI solutions. But unless there are actual ‘learning algorithms/functions from data’ involved, the solution is not an AI solution. This is the source of some of the confusion about what AI actually is.

How to identify a problem, and build a case for AI as a solution

When implementing AI into your business, you have to start off with being very specific on which problem you want AI to solve.

“First you identify which problem, you want AI to solve. It can be a good idea to start off with solving a problem you have already solved, to see if AI can optimize your solution. This way you choose problems and solutions you are already familiar with, and then take it one step at a time from here. It is important that you know what AI is, and what it can be used for,” Lars Ritland says and explains how he is implementing AI at Delivery Management software company Consignor:

“Currently, I am investigating how Consignor can use AI and optimize in terms of quality, efficiency, maintainability etc.. The goal is to provide an even better and more accurate service to the customers.”

Artificial intelligence has become increasingly widespread in almost all industries during the last decade and it is not going to slow down in the future.

“AI is here to stay, but how AI is used will change in the future. Today AI is used to solve specific problems. In the future we will see a more general type of AI that is able to learn more and solve more complex tasks,” Lars Ritland concludes.

Read more: 6 technologies that will change the logistics industry by 2030

 

DAF semi-autonomous trucks

DHL will in collaboration with DAF test semi-autonomous truck platoons on UK motorways in 2019. Source: DAF.

Facts

  • Artificial Intelligence dates back to 1959.
  • AI is defined by Gartner as a collection of data science techniques that seek to mimic human reasoning.
  • AI (machine learning) analyzes in- and output in a dataset and comes up with an algorithm to predict outputs of input data of similar, future datasets. In other words, with AI you learn the algorithm from data, while in conventional problem solving, you program the algorithm.
  • DHL has announced that they want to reinvent themselves with AI in the form of semi-autonomous truck platoons, chat-bots, and fraud detection cameras (1).
  • Netflix, amongst other streaming services, use AI to recommend movies and series based on what the user has previously seen.

 

Text by: Consignor, news@consignor.com

Navigation