Have you ever tried to fix your car yourself, or to give it to an auto repair shop? If yes, then you’ve gone through the tedious process of ordering spare parts. You’re never really happy leaving your car with a mechanic, because you know it always takes too much time to be fixed. You would be surprised to learn that most of that time is spent looking for spare parts.
Last week, my car candles were not igniting, leaving me carless for a week. What I faced wasn’t the difficulty of changing the candles, but rather finding the right ones at a fair price.
The process of finding and ordering spare parts is done through calling suppliers or sorting suppliers databases until you find the ones that suit your needs. For each part, there are many options (manufacturer approved, independently made, etc) all the way to the materials used. When you’ve at last found what you’re looking for, negotiation and order start, and they can be a lengthy process taking up to a few days until shipping.
Thanks to natural language processing, chatbots built with artificial intelligence have the capabilities to offer customer support and resourceful assistance to make your experience much easier. Chatbots go as far as understanding images and recognizing models, parts or vehicles specification. Through image recognition, they can make product recommendations and simplify the product education in the sales process cycle. It is impressive to see that some chatbots have drastically improved, to the point of passing the Turing test.
What can chatbots and image recognition deliver?
Imagine coming across an automobile expert bot. Send it an image of your car. Capable of recognizing the model of your car, the bot is now aware that every car part you are uploading is related to this specific model. Technologically this is very well explained in a paper by Springer, if you want to learn how image recognition can recognize vehicule styles.
Then, through a centralized database, the chatbot is capable of identifying your missing parts, their quality and price. There are more than 20 manufacturers of car candles for a Golf 6 2010, all with different prices. As the person ordering, you would be able to choose and pick the ones you want, based on the selection the bot just sent you. This significantly reduces the amount of time needed to search from different suppliers and figuring out which price is the best and why. Imagga has a use case with KIA that can help you understand further the capabilities of image recognition.
And what if the car mechanic has never seen this car before and needs to look at some kind of documentation before jumping in? In this case, mechanics use car schematics for specific parts of the car. Schematics can be difficult to find sometimes and often not very detailed. Through chatbot technology you can upload the image of your car and request any blueprints you desire using image recognition technology. This can also be done through simple text analytics “I need the blueprints for a Golf 6 2010”.
And the wonders don’t stop here! There are many more collaborations of chatbots and image recognition: student driver bot, car maintenance, tire analyser, etc. If you have other ideas, leave us comments under this article!
Can I integrate my chatbot in Facebook Messenger?
Facebook Messenger is the go-to platform to build chatbots today: they provide loads of visual elements to make your bots more than simple text, and truly put bots on the frontline. But others can be relevant to your target! Slack, Kik, Skype, Twitter, etc. SAP Conversational AI provides a tool that allows you to easily connect your bot to any channel: Bot Connector! Integrated in the SAP Conversational AI bot building platform, or as a standalone software, it’s your best shot of easily integrating bots in Messenger! So, follow this tutorial to setup Bot Connector.
If you are in the automobile industry and got inspired, take your first steps with SAP Conversational AI and Imagga. This is a use case that can be applicable if you are a parts manufacturer, auction reseller, car market place, car manufacturer etc. And even if you have another use case in mind, don’t limit yourself: there can be a bot for everything with a free image recognition API and bot building technology.