AI Technology in After Sales - 4 Use Cases

AI-based technologies and automations are revolutionizing many business areas. However, the designs, maturity and range of potential uses for artificial intelligence vary greatly by business unit. In sales and after-sales, personal interaction and interpersonal relationships still play a major role, but there are aspects where AI technologies and automations can provide valuable support.

Great potential lies in machine, plant, and vehicle construction in the networking of machines, plants, and equipment as well as in the processing of operating data. Here, it is important to evaluate possible areas of application for AI-based technologies and review them with regard to possible implementation.

In this blog post, we look at 4 use cases of AI technologies in after-sales and service, which are being tested for implementation today or have already been deployed:

  • Use case 1: Spare parts identification with AI-based technologies
  • Use case 2: Predictive maintenance and identification of upcoming maintenance needs
  • Use case 3: Customer self-service portal with intelligent chatbot for after-sales support
  • Use case 4: Intelligent linking of service-relevant information using knowledge graphs
  • Objectives & Challenges of using AI technologies in after-sales
  • Conclusion: Cross-departmental exchange as the basis

Spare parts identification with AI-based technologies

Use case 1

A service technician is called to a defective machine and needs to repair it as quickly as possible. But which is the correct spare part he needs to replace? In addition to experience and personal knowledge, AI-based technologies can provide support here. This is made possible, for example, by spare part recognition via photo: the service technician takes a photo of the removed part with his smartphone or tablet and sends the photo to a service. This service accesses a cloud-based, neural, trained network that was created based on all relevant spare parts information from the manufacturer of the defective machine. The service retrieves relevant results for the spare part with details of the probability of a hit.

The results are displayed directly in the digital spare parts catalog or service information system. By clicking on one of the results, the service technician jumps to the corresponding part in the spare parts catalog. Here he receives further information on the part, which is directly linked to relevant sections from the technical documentation and to the ordering option.

The final selection of the correct spare part is still the responsibility of the service technician. However, he is significantly supported in this by the narrowed down results and the specification of the hit probability. In addition, the service technician needs less knowledge about the parts and components. Another advantage is that with every use of the AI-based service in the field, the AI learns, becomes more intelligent over time, and the quality of the results improves.


Spare parts recognition via photo sounds interesting and you would like to learn more about it?

We are working on this exciting topic together with Bosch Cognitive Services. Together, we offer a full-service model consisting of the digitization of spare parts data, the training of artificial intelligence, ongoing service and complete integration into the spare parts catalog and service information system.

Predictive maintenance and identification of upcoming maintenance needs

Use case 2

A typical use case of AI technologies and machine learning in service and after-sales is predictive maintenance. The basis for predictive maintenance is the permanent monitoring and evaluation of machine and process data. Real-time analyses in combination with big data are used to determine the current status of machines, systems and devices in operation. In combination with other parameters and information, algorithms are used to predict future maintenance requirements and the optimal time for performing them. Ideally, maintenance is carried out before a malfunction occurs, but only if it is actually necessary.

Whether predictive maintenance is the right maintenance strategy for a company depends on various factors. You can find more information about this in our blog post "Predictive Maintenance - Definition, Requirements and Advantages".

A practical example of the use of artificial intelligence to identify upcoming maintenance needs can be found at ŜKODA AUTO: The After-Sales and ŠKODA AUTO DigiLab departments say they are testing the new smartphone app "Sound Analyser". This app supports service technicians in quickly and precisely identifying a vehicle's upcoming maintenance and repair needs with the help of AI technologies. By recording operating sounds and comparing them with stored sound patterns, the app determines whether there are any discrepancies.

Once discrepancies are identified, an algorithm analyzes what might be causing these discrepancies and how they could be remedied. The goal is to increase efficiency in maintenance, shorten workshop visits and thus increase customer satisfaction. In 2020, the app was tested in practice by 245 ŜKODA dealers in 14 countries.

Customer self-service portal with intelligent chatbot for after-sales support

Use case 3

Less complex, but no less interesting, is the use of a digital customer portal with an intelligent chatbot that provides support in after-sales and service. The AI-supported chatbot or intelligent virtual assistant does not serve as a replacement, but rather as a supplement to "human" customer service. Common areas of use are processing routine inquiries, bridging the waiting time until a “real” service employee is available, or querying required data.

Advantages of AI-powered chatbots and intelligent virtual assistants:

  • 24/7 availability: regardless of service hours, holidays, and weekends
  • High processing speed and efficiency
  • Relief of human customer service: time-consuming but simple routine tasks are taken over by chatbots

Like all AI-powered technologies, a chatbot only gets smarter with time and ongoing training. Nevertheless, it can already provide support at the beginning of the deployment if it uses a decision tree to answer standard textual questions.

Intelligent linking of service-relevant information using knowledge graphs

Use case 4

According to the German B2B E-Business Report on the topic of "13 levers for a convincing after-sales service", 98% of the companies surveyed are pursuing the goal of networking different systems and software. The status quo often looks like this: Product information is distributed across different systems that are not networked and do not exchange data. Silo thinking is also widespread in the organization: The engineering department does not talk to after-sales service, and technical documentation also works in isolation to the greatest possible extent. As a result, service technicians, service hotline staff and customers themselves need to know which systems contain which information and how they can access it.

However, it can be so much easier: with a service information system that brings together mechanical, electronic, and pneumatic service information. In addition, the system intelligently integrates, links and networks other content, such as product information, technical documentation, supplier documentation, photos, or media assets. The networking and linking of the information can be done based on AI technologies with the help of semantics and knowledge graphs, which build a relationship network of the information.

The result is ONE system through which the various target groups (for example service technicians, support staff, customers, maintenance staff, etc.) can access exactly the information they need for their use case at that time with the help of a simple search option.

Your service information system thus becomes a data hub, spare parts catalog, web shop, sales tool, and internal information system - a "Google" for your service data, so to speak.

Objectives & Challenges of using AI technolgies

Objectives

Technologies based on artificial intelligence are used in after-sales and service with the following objectives:

  • Improving service quality
  • Increasing efficiency
  • Saving costs
  • Optimizing processes
  • Improving scalability
  • Offering new services
  • Enhancing reputation
  • Increasing customer satisfaction

 

Challenges

A technology based on AI is only as intelligent as humans allow it to be. The quality of the data is critical. After the initial training, the algorithm needs a lot more data from the field to learn bit by bit and become increasingly intelligent.

Conclusion

Interdepartmental exchange is the basis for targeted use of AI in after-sales

As things stand, not all AI technologies are yet mature enough to be used in practice in a targeted manner. And not every technology is suitable for every company. To identify meaningful starting points for suitable AI solutions for the company's individual situation, the following aspects are essential:

  • Interdepartmental exchange
  • A thorough analysis of needs
  • Combination of specialist knowledge and IT expertise
  • Merging of relevant data
     

Service is the new sales. Companies that have recognized and taken this development into account have a good chance of staying ahead of their competitors and generating growth even in a competitive market.