The digital economy is one of the most important drivers of economic growth and competitiveness today. Across markets, a range of digital platforms and devices allow consumers to access just about any product or service from the comfort of their own home. Exacerbated by the ubiquity of mobile devices and mobile applications, e-commerce sites have proliferated in the last decade. It is not uncommon nowadays to find at least two local e-commerce companies in each market, in addition to the usual international giants.
Chinese Taipei’s Momo.com Inc (MOMO) has steadily grown in the digital space since first providing mail-order purchases in 2005. As of today, it is among Chinese Taipei’s largest business-to-consumer (B2C) platforms, offering everything from beauty and household products to home and travel insurance.
Like other e-commerce platforms, the rise of MOMO is largely due to the widespread digitization of human activity, generating the kind of massive data sets necessary to know and understand their customers. Browsing habits, purchase preferences, transaction history – a range of data points need to be captured behind the scenes so that digital platforms can improve and personalize customer experiences.
In the case of MOMO, data is the invaluable resource that allows it to efficiently meet consumer demand for quick delivery of purchases amidst increasingly complex supply chains.
Using AI for a unique view on patterns and trends
Since 2017, MOMO uses AI algorithms and data analytics to improve its understanding of patterns in customers’ purchasing habits, of intertwined supply and demand factors, and of the optimal allocation of logistical resources.
MOMO’s Smart Logistic System aggregates data points, taking demand factors – (i) product brand and type; (ii) consumption volumes and frequency; and (iii) customer demographics and location – into consideration with supply factors – (i) warehouse capacity; (ii) inventory levels; (iii) marketing events; and (iv) delivery schedules.
This exercise allows it to model patterns in supply and demand, effectively predicting future shopping behaviors and purchasing trends. This gives MOMO a unique edge, as it can streamline the entirety of its logistics processes for an intelligent logistics delivery system – thus enabling quick and effective response to customer demand.
Making logistics more agile and resilient
The AI system helps improve four key aspects of MOMO’s business operations:
- Creating the optimal warehouse product configuration to reduce delivery times
Developed to facilitate express delivery, the AI algorithm optimizes warehouse inventory based on product classification, vendor, brand, and other attributes – while also taking into consideration warehouse capacity and transportation schedules – for an efficient and flexible allocation of vendors’ products to satellite warehouses in different locations, at optimized volumes.
- Multiple data points allow for informed stock-replenishment decisions
MOMO’s AI-based replenishment model analyzes customers’ orders, their profiles, and online behavior, as well as information on the company’s own inventory levels and shopping events, to generate recommendations to vendors on when and where to replenish stock. This allows them to send their products directly to the satellite warehouses that need replenishing, and reduces routing costs and optimizes inventory management with fewer items needing to be housed in the main storage facility.
- Strengthening warehouses’ ability to respond to temporary surges in demand
Sales, promotions, and other marketing events on MOMO’s platforms can lead to temporary surges in customer demand. Keeping track of and responding to these events on their own can be challenging for warehouse and delivery systems. MOMO’s real-time ‘hot sales’ prediction model uses sales forecasts, inventory changes, and delivery volumes to help main warehouses prepare for surges and adjust inventory levels and schedule distribution to satellite warehouses in advance.
- Achieving cost-efficiency in last-mile delivery
MOMO has deployed numerous small satellite warehouses across the island. These are strategically located in the periphery of major customer shipping locations. However, to streamline last-mile delivery from satellite warehouses to the customer, MOMO has created a ‘satellite warehouse logistic and distribution area analysis module’ to determine the most cost-effective delivery route to use for express delivery. Using customer addresses, popular delivery locations, and distribution area simulations, the AI algorithm generates the shortest delivery route possible and recommends it to MOMO’s delivery drivers.
AI adoption can mitigate against future supply chain disruptions
The impact of adopting this smart warehouse system has been substantial. On the frontlines, MOMO express delivery has reduced delivery time to less than 6 hours, limited overhead costs, and increased customer satisfaction. Behind the scenes, using AI to determine how inventory should be best allocated across its satellite warehouses has allowed MOMO increase its express delivery volume by 500%.
These are major benefits in and of themselves, but they have proven especially invaluable during the COVID-19 crisis. The empty grocery shelves and medical supply shortages witnessed the world over highlighted the importance of smart supply chain management that not only responds to sudden demand surges, but also takes proactive steps to suggest the best measures to take in advance.
Predictive AI, which can help retail and logistics industries anticipate which items would be needed during a crisis, is one example of an AI application that will help businesses prioritize the supply of certain items and reduce the chain effects and other negative consequences of panic buying.
MOMO credits its ability to cope with the unanticipated COVID-19 led spike in consumption to its smart warehousing system. Demand for large volumes of toilet paper, which are bulky and ties up logistics capacity, would have restricted movement of other anti-epidemic products and necessities. Using predictive AI, MOMO was able to distribute toilet paper directly to its satellite warehouses, which could then be quickly delivered by last-mile delivery, eliminating the need to store it while waiting for transport schedules between main and satellite warehouses to match up, and preventing orders of toilet paper from jamming the flow of other goods.
In this context, the use of AI in logistic networks can address supply chain vulnerabilities and help prepare for future disruption, thus supporting broader economic dynamism and resilience.