Know It All Nev

On Demand Reviews – 2021

Category: machine learning

How Blocking the Suez Might Dampen Summer time Demand

Nifty Nev
artificial intelligence, automation, demand, demand forecasting, Demand forecasting in supply chain, disruption, forecasting, machine learning, probability, resilience, Retail, Supply Chain, supply chain disruption |

In what feels like an anti-climactic end to an incredible story, the Ever Given – the boat that broke the internet when it blocked the Suez Canal back in March – has finally docked at Felixstowe, England, with its cargo from the Far East – a full four months after it was scheduled to arrive. 

It was one of those great black swan events that kept many of us enthralled for days, but what can this whole event teach us about supply chain resilience? Billions of pixels of digital news have been used to describe how supply chains need to understand the macro picture, like shipping delays and supply impacts. Today, I want to focus on the micro. I want to follow the ripples of the Suez Canal situation all the way to the UK high street.

In an earlier blog, I wrote about volatile customer demand and how a modern machine learning forecasting algorithm needs to account for uncertainty. To re-cap, the Blue Yonder demand forecasting service looks for relationships between influencing factors, such as weather, price, and season, rather than attempting to layer on top of a demand forecast based on historical sales. That is the key to understanding uncertainty and driving better forecast accuracy. Now that the Ever Given’s cargo has finally landed in the UK, there is a lot of uncertainty about to disembark.

Uncertain Supply

The Ever Given holds around 18,000 shipping containers containing a mix of perishables and non-perishables. The perishables will be destroyed, but many of the non-perishables will have been timed for start of spring. Some may go directly into UK market via discount retailers or re-directed to markets further afield.  For what remains in the UK market, retailers are faced with inventory that is out of sync with the season.  Couple this with the increased cost of shipping caused by the pandemic and you’ve got the perfect storm of tight margins, increasing prices and too much stock.

Uncertain Demand

Based on historical data, we know that customer sensitivity to price change is not uniform, either geographically or across the year. A price change for T-shirts in mid-summer may not have the same effect on demand in spring, unless there’s also an unseasonal heatwave. The location of the store also plays a critical role in this relationship, with coastal stores typically seeing higher traffic during the summer. 

This is far from a typical UK summer though. Temperatures have been hotter and drier in the north, while the south has recently seen storms. Meanwhile, with COVID-19 restrictions on travel and changing regulations, more people are staying at home for holidays.

Many of the items on the Ever Given were originally due to be sold at the end of spring. Seasonal items are typically discounted at the end of the season, but with higher shipping costs, increasing retail prices and potentially more local competition on the high street, you need to understand how your customers are likely to behave, while being ready to pivot when they surprise you by shifting purchases back to stores, or looking for outfits to hit the town again.

Blue Yonder Forecasts Better

Blue Yonder can help automate your demand forecasting with its granular, proprietary forecasting service designed to self-learn trends at local levels and dynamically respond to the crazy weather or changing store patterns from the stay-cation craze of 2021. A demand forecast that understands complex uncertainty but simplify human control can help build better allocations, improve resourcing and ultimately feed into better pricing decisions. It will help your business teams navigate the uncertainty that surrounds us every day, and lead to improved business KPIs.

Black swan events like the Suez Canal blockage will always occur, and tools like Luminate Control Tower can help us better manage them. We cannot reasonably predict black swans, nor can we learn anything meaningful from them for future forecasts. It is the daily uncertainty that we need to manage; and if the management of day-to-day disruption is automated, planners should have more time to deal with the black swan events when they occur.

Like the slow progress of the Ever Given out of the Suez Canal, the effects of uncertainty can take a long time to manifest in the market. With the lived human experience of pandemics in short supply, highly intelligent and automated demand forecasts are the critical foundation of a resilient supply chain. Major shocks get most of the attention, but markets are uncertain every day. Sometimes because of weather, sometimes because of price, generally because of the combination of all the factors operating together in a connected network at the local level.

If the future of retailing continues to be uncertain, the heart of your supply chain needs to do more than cope. It needs to thrive by embracing uncertainty.  Blue Yonder’s unique approach has been designed to deliver the supply chain of tomorrow. Please contact us if you’d like to discuss how we can help transform your business.

Read More

An Clever Technique to Enhance Demand Forecasting

Nifty Nev
accuracy, artificial intelligence, automation, demand, demand forecasting, forecasting, machine learning, probability, Retail, Supply Chain |

I’m a cyclist. But I have a confession to make. I’m a fair-weather cyclist. While riding my bike helps keep me healthy and gives me time to think, when I get caught in a rain shower, I know I’ll be spending a significant chunk of my afternoon cleaning mud and gunk off my bike. And so, based on my willingness to accept risk that day, I make a judgement call. Before I leave the house, I always consult the weather app on my phone. Anything more than a 30% chance of rain and I stay inside.  I estimate that has saved me hours of de-greasing in the past year alone.  

I find it interesting that concepts that we readily accept with our personal devices don’t always seem to resonate in the workplace.  When I’m considering whether to risk cleaning my bike after a ride, I rely on a weather forecast that may or may not materialise. I don’t question whether the forecast is right, I accept that it is uncertain; sometimes it rains, sometimes it doesn’t. I make my decision based on a preference for staying dry.  

Demand Forecasts are Uncertain 

The same principles of risk and uncertainty that we accept in weather forecasts apply equally to demand forecasts. Order too much and you lock up capital in slow moving inventory or eat into margins for perishables Order too little and you run out of stock and disappoint customers. Planning inventory relies on an accurate demand forecast. Understanding that forecasts always include a level of uncertainty is the critical first step toward resilient supply chain automation.  

So how do you build a resilient demand forecast that understands uncertainty? Traditional approaches to forecasting are fundamentally flawed. Basing a demand forecast on sales history alone doesn’t measure true demand. Local out-of-stocks are written into the sales record alongside historical weather patterns and events, creating future manual work for demand planners, such as holiday and cultural events like Easter, or trying to “correct” last year’s wet-and-washed-out summer. This problem is compounded when daily profiles are applied to a baseline forecast before manual adjustments are made. 

Weather only repeats itself 15% of the time, but it’s often used as an influencing factor for demand.  We shouldn’t be surprised that demand planners spend a considerable amount of their time tuning algorithms then adjusting the output to something that looks more “correct,” based on their experience. The problem with experience, especially in today’s fast changing market, is that no-one really has relevant experience that they can draw on.  

McKinsey research tells us that the past is no longer a guide to future behaviour, as consumers increasingly shift channels and are looking for new brands and more convenient experiences. Forecasting capabilities that look backward to predict the future are fundamentally flawed. They struggle to adapt to changing shopper behaviour and cannot measure uncertainty without making assumptions. They leave demand planners exposed to inventory risk. 

Embrace Uncertainty in Demand Forecasting 

Blue Yonder corrects these flaws by forecasting via a unique and more accurate approach. Our demand forecasts are built on machine learning analysis of the relationships between many different historical data sources, such as weather, special events and price. We don’t use artificial intelligence to layer influencing factors on top of a baseline, but rather look at how strong each influencing factor is at any given point in time and use this as the basis for forecasting into the future. Missed sales are accounted for and special events are automatically moved as calendar dates shift. 

Once trained, the forecast engine tests itself against actual sales every day to ensure that as customer behaviour changes, the model is able to self-correct. Configurations like weekly profiles and assumed statistical distributions are made redundant as the forecast is created every day from the ground up, using the most recent data to improve accuracy. This changes the role of demand planner from algorithm tuner into data custodian and strategic advisor. 

Our unique demand forecast predicts the full spectrum of demand, not just the mean, and it calculates the probability of every unit of demand at the item, store, day level rather than assume a shape or profile. This information becomes valuable when you later want to make inventory decisions, recommend price changes or make assortment changes. Irrespective of how close your mean might be to historical sales, there is always a chance that customers will want to buy more or less of what you predict.  

A demand forecast that intuitively understands what factors drive shopper behaviour is useful for predicting demand for all retail assortments. Fresh produce demand can vary widely across the week as the weather changes. Promotional demand varies depending on pay day, local competition and the season. And, of course, seasonal products like BBQ accessories and clothing can depend disproportionality on weather. Chasing volatility manually cannot be accomplished with the precision required to make a difference at the point of purchase. Planners might plan at a cluster level, but shoppers act locally.  

Accurately understanding this complexity at the local level is extremely useful for creating better demand predictions but automation is not possible without trust between humans and the machines that serve them. The complex reality captured by our forecast engine is presented to humans in a meaningful way, with data grouped in buckets that make sense in plain language. Weather, promotion or trend rather than temperature or shelf position, breaking the black box problem that plagues most machine learning solutions.  

By adopting an automated, self-learning demand forecast like Blue Yonder’s, planners no longer need to tune algorithms or adjust forecasts to account for changing weather or promotional calendars. The forecast can be largely automated, leaving planners to focus on system-generated exceptions, and collaborating with others on how to best use more intelligent and accurate forecast to drive better results, like getting the right amount of inventory where it is shoppers are likely to want it.  

This re-imagined approach is all very simple to the user. A lot like my Sunday afternoon when I can skip giving the bike a bath and focus on more productive things.  

Read More