The digital wave has swept every industry and logistics is no different. The rise of eCommerce, intense competition in industries like retail, eCommerce, FMCG, hospitality, and healthcare etc. around the world has given consumers plenty of convenient buying options at competitive price. Customers want to receive their parcels at the earliest, once they have clicked the order button.
In a parallel world, the manager wants to know about every parcel on real time basis. For the manager, the visibility is no longer about “where is my truck?” It is now about “where is the parcel?” It is being drilled down to SKU level and coupled with predictive analytics is changing the game.
Given that, last mile delivery forms such an important part of the buying experience, brands have no choice but to remodel their logistics functions. Therefore, they are increasingly leveraging the power of artificial intelligence and machine learning to improve their logistics decision making.
Artificial Intelligence is enabling predictive logistics and redefining the user experience in many ways. Few of the operational use cases are listed below:
Efficient Job Allocation- The fundamental problem in last mile logistics is known as- Travelling Salesman Problem (TSP). It is about identifying the shortest route a salesman should take to visit a number of locations and return to starting point. If there are ‘n’ number of locations to be visited by a delivery person, there could be 1x2x3…x n possible ways to do it. The problem can get even more astounding if there are hundreds of delivery personnel present at different locations. The manager would find it very hard to deep dive into this pool of data and allocate jobs to the team based on skills set, traffic conditions, and impromptu demand. However, a mobile-based logistics management solution can perform this task efficiently using geo-intelligence for route planning and location of the delivery person. Additionally, it can also be customized to make analytical decisions based on factors like skill sets, dynamic demand, and return/reschedule requests from the customers. Geo-intelligence also enables the system to automatically calculate the expected time of arrival (ETA) and share it with the customer in real time. It sends proactive notifications to the delivery person, manager, as well as customer in case the delivery is expected to be delayed due to any contingent reasons like traffic jam or route deviation etc. Therefore, it brings the much-needed visibility to the logistics industry along with an added layer of visibility and predictability.
Machine Learning based geocoding – Enterprises must use sophisticated algorithms to self-learn the ‘correct’ address from listed delivery addresses and their success and failures. Machine Learning based geocoding platform searches from its internal billion plus addresses, breaks the address into multiple parts to increase the hit from map providers and uses different map providers to increase the geocoding accuracy. The addresses where deliveries have been made in the past and the same information can be used for future deliveries to not only the same person, but even the same building. This results in best routes and quicker deliveries.
Additionally, the algorithms also lets the application schedule deliveries basis ‘type’ of address – home or office. For instance, a delivery to an office will not be scheduled over the weekend or on public holiday. Error free planning, keeping multiple scenarios in consideration not only reduces the man-hours for logistics company, but improves the overall delivery happiness score.
Customer Support-At any stage of buying cycle, customers may face troubles. Logistics companies are deploying self-learning virtual customer assistants to interact with customers to address simple and repetitive troubles or queries. The system can recognize speech as well as data (customer, contract, transactional, and operational information) to provide human-like experience to the customers who need help. Artificial intelligence enables the system to consider multiple scenarios within a couple of seconds (something completely impossible for human support staff) and provide the best possible resolution. The system also learns through new problem scenarios shared by customers and provides solutions faster if the same problem comes again in the future. By deploying such artificially intelligent system, logistics companies can save their human resource for more complex cases, while improving the customer experience manifold.
The world today is turning towards Blockchain, Machine Learning and Artificial Intelligence as it transforms the existing business models. It is an undeniable fact that these technologies would help automate and streamline logistics processes. But are the businesses are ready for the change with minimal disruption?
By: Kushal Nahata, Co-Founder and CEO, FarEye
