Leveraging Advanced Analytics for Demand Management : A Retail perspective

The importance of Demand forecast

Supply chain challenges have evolved exponentially, become more. Many organizations are adopting a Demand Driven supply chain model, which involves building supply chains that work in response to customer demands. Because Demand is the driving force, Demand management becomes very critical in this model.

Due to aspects like global operations, shortened product cycles, and more volatile business environments, the management of uncertainties is one priority for the Demand driven supply chain and hence Demand forecasting becomes a very essential exercise. Used primarily to forecast sales, these forecasts are used in multiple ways. Beside production planning, inventory management, market entry strategies, and analysis of customer behavior, different demand forecasting methods allow for prediction of probable scenarios based on historical data and prevailing trends.

A precise demand forecast provides an accurate picture of future demand and helps to avoid overproduction and excessive overstock. Although forecasting methods allow quantification of future demand, supply chain uncertainty refers to situations in which the decision maker acts under a set of different forces. These forces can be further specified to understand the relationships and effects that play a strong role in planning, improving efficiency, or obtaining accurate forecasts.

The Demand influencers

The underlying core for better Demand management is to first get a grasp on what are the key factors that influence Demand. Figure below illustrates an overview of the factors influencing demand. These factors are embedded into a customer-oriented supply chain. Uncertainties in demand forecasting depend on the objective of planning. The state of a product, determined by the “four Ps” in the marketing field, has an important relation to the underlying goal of fulfilling a customer request. Important parameters for demand forecasting include the product (ex: for identification of future demand), placement (ex: for determining the modes of transport), pricing (ex: for implementation of a new product), and promotion (ex: for determining the target market).

Capture

Figure 1: A high level overview of Demand Influencers

In addition, the supply chain is associated with constraints that directly influence the forecasting process. The supply side as well as the demand side not only affect how information is processed but also define the sufficiency and quality of information. Finally, the environment must be considered in demand forecasts as the role of competitors and external factors, such as weather conditions or special events, affect the efficacy of control actions.

To summarize, the following are the key factors that influence Demand. Consideration of all these elements is needed to determine the appropriate demand forecasting method.

  • Products
  • Consumer preferences
  • External factors
  • Marketing factors
  • Retail factors
  • Supply Factors
Forecasting Method Data Points Challenges
Grassroot Forecasting (Sales force composite) Customer’s intention to buy in the near future, Understanding of and relationship with the customer Laborious in retail situations and vague results
Market Research Customer’s preferences and insights data Diverse data required to further improve results, need for identifying trends earlier and quantifying the value of marketing measures
Expert forecast estimation Domain knowledge, data foundation for decomposed decisions, quantitative analogies Difficult to test the expert’s hypothesis, pure data-based decisions not possible
Time series forecast Historic sales figures, Influence of different seasons, Understanding of the development of a trend Requires more recent data, needs to be interrupted when special events occur
Causal demand forecast Knowledge about the distinct factors influencing demand, Large datasets, Theoretical background, Knowledge of causal relationships, Segmentation strategies Only a few causal relationships are known with certainity and testing new ones are challenging, requires recent data and relatively expensive to produce forecast

Table 1: An overview of commonly used forecasting methods

An overview of Advanced Analytics (AA) techniques for Demand Management

Advanced analytics (AA) addresses complex business questions, combining methods from statistics, data mining, and machine learning. AA is differentiated from simple analytics by the types of questions that can be answered with the computer model; simple analytics is associated with explicitly formulated questions that can be answered by a few SQL queries, while AA is data-driven and aims to reveal insights and answer implicit questions. The output of AA is computer models that can automatically process data, process data. There are categories of AA and for our purposes, we will focus on the following three categories:

  • Descriptive (DAA)
  • Predictive (PDAA), and
  • Prescriptive analytics (PSAA)

DAA, includes computer models that use clustering, association rules, and classifications. It also involves the process of preparing data sets. To be feasible inputs for computer models, diverse data sources need to be structured. Deep neural networks and text analytics are two of the most important types of data processing. Commonly applied DAA techniques are cluster analysis, principle component analysis, association rules, classification and decision trees, ensemble learning, logistic regression and neural networks, and naïve Bayes classifiers.

PDAA, includes estimations. Estimation models are more difficult to create than classification models; whereas classification models have to predict only a few values (nominal, discrete target), continuous estimations must predict every value contained in the target variable (metric target). In addition to classical regression analysis, other methods are often applied, such as time-series models, regression trees, and artificial neural networks. The most appropriate methods for forecast demand, linear regression, and time-series analysis are described below. Artificial neural networks do not suit this endeavor because the model entails black-box elements.

PSAA, is a type of applied learning as it uses data obtained from descriptive, predictive, and domain knowledge and applies it to real cases. In this way, the system determines the best situation-specific action. PSAA is an emerging field involving advanced optimization and simulation as well as game theory and decision analysis methods.

Demand Forecasting in retail Supply Chains

To illustrate the application of Advanced Analytics (AA) for Demand forecasting management, we will use the example of a retail supply chain. One primary reason to chose retail is due to the complexities involved in retail forecasting process.

If you have not been exposed to retail forecasting, you may view retail demand forecasting as straightforward time-series forecasts. However, these are only applicable in repetitive and short-term situations, and most retailers face varying product demand due to diverse factors, such as weather or short-lived trends. At a product’s launch, retailers often possess no sales experiences but the product performs well due to promotions, the opening of new stores, or changing of the product assortment.

All of these determinants need to be identified and structured to acquire a profound understanding of demand challenges. Based on my experience, following are the key influence factors for six key demand forecast categories. In order to effectively manage demand via Advanced analytics, it is critical to ensure that these influencing factors will be addressed by those analytical methods.

Demand forecast categories Influence factors
Products Quality
Competitors, substitutes, complements
Prices
Consumer preferences Fashion and trends
Buying behavior
Brand awareness and perception
External factors Weather
Special events
Seasonality
Income level, economic outlook
Local development
Mega trends (Demographics, technology, climate etc.)
Marketing factors Promotions
Advertisment
Retail factors Local competition
Shop attractiveness
Shop assortment and layout
Supply Factors Available products at point of sale
Expiring products

Table 2: Key factors that influence demand in retail Supply Chains

Leveraging Advanced Analytics in Retail Demand forecasting

Now that we understand what the key categories and influencing factors are, let us apply the three Advanced Analytics (AA) methods discussed in the previous section. Note that rather than define exactly what type of Analytical method can be used, I will be describing the analysis that needs to be done, for each influencing factor and analytics type. Having this understanding and having familiarity with tools that fall under the three types, you can determine which tool will be a good fit for your unique requirements.

Product and Consumer preferences

Getting an in-depth grasp on how product characteristics and product ecosystem impacts demand is one of the primary aspects of demand management. Same goes for consumer preferences. Together- Products and consumer preferences influencing factors are the most critical factors and hence careful selection of analytical approaches is extremely important. The table below summarizes the analytical approaches that can be leveraged for each influencing factors in these two areas. Note the full form of the abbreviations used in the tables are:

  • Descriptive (DAA)
  • Predictive (PDAA), and
  • Prescriptive analytics (PSAA)
Demand forecast categories Influence factors Descriptive (DAA) Predictive (PDAA) Prescriptive (PSAA)
Products Quality Assess product quality based upon reviews, customer feedback, warrantly claims etc. Predict influence from quality factors upon demand for a product Optimize assortment with the best price quality attributes
Competitors, substitutes, complements Build a products network to understand their interrelations Predict cross price elasticity Optimize the price structure of products
Prices Structure the development of prices Predict price elasticity Not applicable
Consumer preferences Fashion and trends Detect social media trends that can influence demand, discover where trends start and how they spread, detect trend reversals Include social media trends in forecast models or test their influence on sales Simulate different product trend scenarios
Buying behavior Segment customers according to their demand, detect purchase behaviors with association rules, create a comprehensive view of customers from different data sources Predictive models based upon individual customers or customer segments, create forecast models with diverse input variables, predict customer visitor traffic Optimize demand forecast models with cost factors
Brand awareness and perception Describe brand awareness and perceptions Predict brand influence on product demand Not applicable

Table 3: AA to manage Product and Consumer preferences influencing factors

External Factors

While product factors and consumer preference factors are critical, demand can be influenced by many external factors that are beyond the span of consumer preferences. These external factors sometimes may override consumer prefernces as well. Owing to this, a careful analysis of analytical approaches in these areas is key to a comprehensive demand management program. The table below illustrates the analytical approaches for these area, for the three categories:

  • Descriptive (DAA)
  • Predictive (PDAA), and
  • Prescriptive analytics (PSAA)
Demand forecast categories Influence factors Descriptive (DAA) Predictive (PDAA) Prescriptive (PSAA)
External factors Weather Categorize weather forecast data Predict how weather influences demand, integrate categorized weather data into the models Optimize assortment according to weather
Special events Categorize events based on certain criteria Predict how event type influences local demand Optimize assortment depending on event type
Seasonality Categorize products according to their seasonal demand Predict the influence of seasonality on demand Optimize assortment depending on season
Income level, economic outlook Categorize economic predictions, understand how experts correlate different indicators of economic development Predict the influence of economy on demand Optimize assortment depending on economic factors
Local development Discover drivers of demand that arise from local development, understand where customers come from Predict how much local and regional drivers influence demand Find interesting new shop locations using location analytics
Mega trends (Demographics, technology, climate etc.) Build a knowledge graph of related long term trends and related technologies to make future development more accessible Not applicable

 

Simulate influence of trends on customer demand

Table 4: AA to manage external factors influencing factors

Marketing, Retail and Supply Factors

How you eventually Market your product and fulfill customer demand are obviously also critical factors in your Demand management strategy. The table below illustrates how analytics can be leveraged in these tow key areas of Marketing and fulfillment, for the following three analytical approaches:

  • Descriptive (DAA)
  • Predictive (PDAA), and
  • Prescriptive analytics (PSAA)
Demand forecast categories Influence factors Descriptive (DAA) Predictive (PDAA) Prescriptive (PSAA)
Marketing factors Promotions Categorize promotion types Predict influence of promotions and discounts Optimize the timing of promotions, prescribe the amount of additional products needed
Advertisement Categorize marketing campaigns, analyze social media reaction to campaigns Predict demand influence of marketing campaigns Prescribe how many more products shall be stocked
Retail factors Local competition Create a network of local competitors, categorize new entrants and departures Adjust models according to change in the categorization of competitors Simulate influence of new entrants on demand, find locations to expand
Shop attractiveness Rate the attractiveness of retail locations Integrate shop ratings into model estimations Calculate optimal time to invest in shop attractiveness
Shop assortment and layout Categorize products depending on their location in the shop, discover how much time customers spend in the shops Change models according to shop assortments, determine influence of product location on demand Optimize shop layout
Supply Factors Available products at point of sale Not applicable Use prepared data about forecast error to improve models, adjust “short time supplies” Optimize Supply function based upon cost of product scarcity and excess stock
Expiring products Not applicable Find drivers for wasted products Not applicable

Table 5: AA to manage marketing, retail and supply influencing factors

Conclusion

The analytical approaches mentioned, as you can see above, complement each other and hence need to be used in tandem. Using Advanced Analytics (AA) effectively for managing the Demand forecast process requires that you leverage all the three types (Descriptive, Predictive and Prescriptive), where applicable. Leveraging these together will ensure that you gain the optimal advantage from your Advanced Analytics journey.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s