HomeStock management and the shopping season
Another year is coming to an end and most of us are starting to write gift lists. December has always generated a lot of revenue for many retail sectors, with sales spread throughout the month as Christmas day approaches. However, this decade has shown a potential shift in the shopping schedule in households around the world, and November may take December's place as the most impactful shopping month. The increasing popularity of events like Black Friday outside the U.S. and Single's Day (traditional in China) around the world which offer massive discounts can be pointed out as the cause of the shift.
In the past, stores only used to offer discounts on their products in the context of a changing scenario, like a new season in fashion or a hardware upgrade in tech. The goal was to remove products from stock and reduce potential losses. But gradually, the spirit of discounts expanded in the store’s portfolio, reaching the point where customers expect to find special buying conditions for all products.
From the perspective of the suppliers, the high shopping demand massively concentrated on a few days has shown the importance of good stock management. When the company expects production capacity not to match demand, it needs to increase stock levels. This may happen when there is a demand peak or even a large maintenance scheduled for the plan. In these cases, many alternatives are possible, such as:
- Open new production shifts
- Outsource the production
- Increase production throughout many periods leading up to the demand
- Combine the previous strategies
Each one of them has a cost to be implemented, logistical constraints, and storage capacity. In addition to all those concepts, the decision-maker still needs to define when to start executing the plan.
That is a genuine optimization opportunity in many sectors, where the intersection of microeconomic concepts and production planning drives the best answer. For instance, multi-period product-allocation mathematical formulations combined with mixed-integer solvers are capable of determining the production level per product per plant that minimizes the sum of operating, stocking, and logistic costs while ensuring the demand is met. Furthermore, extra savings could be achieved by implementing other algorithms that help better positioning products on the warehouse floor or define picking path routes.
For example, let's consider the challenge that Nike faces. In its 2019's fiscal year report [1], we can find that 41% of its $39 billion were made in the US in spite there are only 384 of its 1152 retail stores. Footwear products are manufactured by 112 factories in 12 countries while apparel products by 334 factories in 36 countries, all outside of the US. In a conference call [2], Nike's CEO Mark Parker has said that online sales grew more than 70% in North America for Black Friday alone. With so many manufacturing facility possibilities and demands scattered across the world, how much an optimized decision of product-allocation can have? It is worth noting that this challenge is also present for medium-sized companies in another economic scale.
After seeing that Alibaba's sales on Single's Day have reached $75 billion in November 2020 [3], we all should prepare our gift lists (and production planning) a little earlier to enjoy this time next year and the new era it has brought to us.
[2] Nike's online sales climb 38% in Q2
[3] Singles Day: Alibaba sales blitz rakes in $75 billion as Chinese shake off Covid-19
Co-author Fabio Silva - Consulting Manager at Cassotis Consulting