Moody Davit Report – Moody Davit Report

Introduction: The Moody Davidt Report features the latest in a new guest column related to Voiceback Analytics based in Bengaluru, India. Voiceback Analytics combines his three diverse areas of expertise: market research, operations and technology to solve businesses. Solve problems and find your way to growth.

“Macbeth! Macbeth! Macbeth! Beware Macduff. Beware the Lord of Fife.”

“No woman born shall harm Macbeth.”

“Macbeth will never be defeated until from the forest of Birnum to the high hill of Dunshinan rise against him.”

(Act 4 Scene 1), “Macbeth” by William Shakespeare

Like Macbeth, we may use witches to predict the future. But anyone who has read Shakespeare’s epics understands perfectly well that while witches foretell the future, they always lead those involved down the wrong path of death and destruction.

Fortunately for us data is not.

data-driven and data driven

As the clouds darken and the water becomes choppy, captains of ships far from the ocean can do nothing but hope for better weather forecasts.

Today’s business environment is volatile and much rougher in some places. Travel retail is one of the sectors most affected by COVID. As the world opens up and demand for travel increases, the resulting interrelated challenges threaten to turn into a perfect storm for the industry. . Skyrocketing air fares due to declining flight capabilities. Lack of Inventory – Lack of shipping containers to move products or packages, or simply goods.

Add to the complication of shrinking GDP and high employment rates in the US market. Even the most famous economists aren’t sure if it’s a slowdown, a recession, or something new?

Solving one problem simply shifts the pressure to a new problem. As growth slows, inventory planning and availability can become the biggest obstacles to increased sales and profitability. Travel retail is still recovering and is still a long way from the peak it achieved in 2019.

Despite reduced customer demand (overall compared to pre-pandemic), product shortages and inventory planning have become key focus areas. To do so, you need the ability to forecast your business six months to a year into the future.

This article describes key technologies and algorithms that can be used to better control future supply.

Importance of inventory

Inventory planning is a rare variable in store profitability that duty free can directly influence and control. Other important variables, such as store rent and foot traffic, are to varying degrees unaffected by the operator.

Here are some scenarios and reasons why inventory planning and purchasing becomes mission critical.

• Purchases happen to match business expectations. Sales forecasts manage brand and product ranges, planograms, and even planned marketing and promotions on the floor.
• Travel retail is a high margin business. The amount a retailer loses on lost sales can be as much as his 50% or more of the selling price.
• The lead time is long due to movement led by sea freight.The product is bulky or heavy and moving it via airfreight can take away a key marketing point
• Customer loyalty is typically brand driven. If the brand is not on the shelves, even for a short period of time, customers may move their purchases to the destination airport, potentially losing them forever.

Data analysis and inventory

A great foundation of modern data analytics is a) the ability to consume data in real time, and b) the ability to process large amounts of data and highly detailed data.

These two can be used effectively to improve the availability of your customer’s product.

average rule

A simple and effective method is to create a predictive model that calculates the average historical sales of each product.

The model creates a calendar that classifies each date based on parameters such as days of the week, holidays, seasonality of the year, special travel events, and promotions. This calendar is used to calculate average sales for each combination based on historical data.

The average of this historical data is used to predict future consumption. Models used in multiple contexts recognize uplift caused by multiple factors such as holiday seasons, New Year celebrations, school holidays, and marketing campaigns. This model relies heavily on considerable computational power and produces continuously updated data for arbitrary time periods.

Traffic forecast

Voiceback Analytics & tRetail Labs leverages our partnership with RateGain (https://rategain.com) to utilize travel ticket data to continuously generate forecasts of arrivals at airline terminals for up to 90 days.

These algorithms have been successfully used for occupancy and rate planning in large hotel chains and are currently being used in the travel retail industry for supply chain planning. They continuously examine macroeconomic parameters such as actual air bookings and political events to generate forecasts of passenger numbers for the coming period.

This forecast is available by city (arrivals) and airline. Combining historical passenger spending by airlines with the dataset allows you to create accurate forecasts for your business over the next 90 days at the product or item level. All past sales data and future forecasts are entered into beautiful intuitive reports that analyze the past and provide accurate revenue and inventory forecasts.

Using queuing theory and the Poisson distribution, I created a model to predict the maximum number that can be sold in a given period of time

Quantity planning

Even after future traffic is shown, buyers still have to decide on the scope and quantity of each product. In high-value categories such as fine watches, fine cosmetics, fine spirits and wines, and fashion, assortment planning is difficult and error-prone.

Buyers must invest capital in inventory and wrestle with the nuanced question of whether to keep 15 or 25 items of a particular watch or a particular cosmetic item. Holding one product more than necessary consumes capital and leads to deterioration over time and a decrease in inventory value. Conversely, maintaining less product than needed leads to out-of-stocks and lost profits.

I used queuing theory and Poisson distribution [probability distribution that is used to show how many times an event is likely to occur over a specified period – Ed] Create a model that predicts the maximum number that can be sold in a given time period.

The model can be run by design, product, or price range and predicts the maximum number of product sales for a month, week, or day. This model can be effectively used to create optimal order volume and shift shelving plans.

This model is based on advanced statistics and can achieve success rates of up to 95%. This model redoes the buying process, reducing both lost sales and unsold sales.

demand sensing

I’ve touched on demand sensing before in this column. Sensing is often used to predict the short-term future based on the short-term past. The concept is based on that if the past is a good way to see the future, then the very near past is a better way to see the very near future. Demand-sensing algorithms can be used to plan one month ahead and match demand with incoming supplies.

Out-of-stock of a major brand at one airport impacts other connected airports and other competing product lines at the originating airport.

road ahead

Inventory planning is a critical component of any travel retail business, and the current situation is not stable for some time. The presence of multiple products throughout the store and the interconnected nature of the business ensures that disruption of one brand affects others.

For example, a stockout of a major brand at one airport impacts other connected airports and other competing product lines at the originating airport. In other words, the goalposts are always in motion, and solving one problem tends to create another. The ability to analyze large amounts of data professionally and quickly is therefore a key determinant of success in a very challenging market.

For more information, please contact manishi@voicebackanalytics.com.