Those of us who love analytics often quote the phrase – Data Drives Decisions. Ok, so why is there so much of fuss around big data? Isn’t it just the same data with a lot of big numbers which needs to be crunched?

Big Data means those datasets where traditional data mining and handling techniques fail to provide business insights into the data in question. It is an overwhelming amount of data generated by all the transactions carried out by an organization. These data sets are so overwhelmingly massive and complex that they were often ignored, or not even stored. Therefore, in today’s technologically advanced world Big data analysis plays an essential role in the supply chain management sector.

Talking about the essential roles of Big data it also has challenges such as how can it be captured, where will it be stored, how can it be visualized, how can one search through it for a specific data, how can one share it or transfer it to others, etc. While these questions are being addressed by recent technological advances such as the Internet of Things (IoT) devices, Cloud Computing, Relational Database Management Systems, Machine Learning(ML), Artificial Intelligence (Ai), etc. Most companies would wonder, is it worth doing all this? Innovative companies such as Google, Amazon, Facebook, etc. have realized that big data is far from being a burden. Big data is a source of competitive advantage by providing a treasure trove of information which gives them invaluable insights into customer behavior and business patterns.

Whether it is about a business, its various functions or the use of its products by customers, data is everywhere. Digitization is hastening the process of capturing every possible opportunity in a process or a transaction as a data point. In the absence of digital touch points, IoT devices play a big part in converting machine related or other such information into digital data points. Big data then is all about harnessing all this data, analyzing it, and converting the data into valuable insights, which can help businesses to be more efficient.

The supply chain domain presents a very interesting area for big data applications due to some key factors:

  1. While Supply chain was earlier treated as a cost center, most organizations are now focusing on it to drive competitive advantage via differentiation and customizations which make them nimbler than their competitors

2. With globalization and changes in the business environment, the effectiveness and efficiencies of SCM are increasingly under the radar

3. Supply chains witnesses hundreds of thousands of transactions which are of varying levels of complexity. These transactions need to be minutely analyzed and optimized intelligently so as to be able to cut costs or reduce lead times or inventory levels, etc.

4. Changing customer expectations via higher service levels are pushing Supply chains to innovate on ways and means of getting their products or services faster to the customer in a more efficient and cost-effective manner.

Some of the key areas within the supply chain management where big data can be analyzed are:

1. Demand forecasting – The Pampers case study has often been quoted to illustrate the devastating effects caused by the Bullwhip effect.  With supply chain managers compensating for delays by vendors or manufacturing, fluctuating demand, and other factors; demand forecasting is a critical activity which affects almost all aspects of an organization from inventory management to resource allocation to financial planning, etc. Demand is also affected by multiple disparate variables, all of which may not have direct correlations or interlinkages. In such a scenario, one can hardly apply traditional forecasting methods to predict future sales.

2. Supply Forecasting – The next critical activity which can majorly benefit from big data analysis is supplied forecasting. When should ordering schedules be released, pickups arranged for, what may be the variability in transportation lead times or custom clearances, how can seasonality be managed optimally, how festive promotion sales can be best supported – answers to these questions are critical for a supply chain manager to avoid a cardinal sin – Being out of Stock.

3. Inventory management – Inventories consume a significant amount of a firm’s resources and present various business risks such as obsolescence, expiry, high inventory carrying costs, etc. With this in mind, a supply chain manager always tries to ascertain what is the optimal inventory holding at an SKU level, what the ideal reorder point is, where the inventory should be stored, etc.

4. Vendor management – Supply forecasting depends highly on vendor reliability. While most organizations mitigate risk by working with several vendors so as to never be out of stock, it is important to understand the variables related to vendor performance which can help in benchmarking performance. Analysing which vendor one should purchase from at any specific time helps in predicting and recommending actions for eliminating/mitigating the effects of any unplanned deviations.

5. Fulfillment & Distribution – Should products be fulfilled directly from factories to stores or via multiple fulfillment centers? What is the right number of fulfillment stores? Should the product movement be via air, surface or sea? What may be the sales mix by product per store per city? What should be the right merchandising mix within a store? What should be the replenishment frequency to each node within one’s distribution? Logistics managers always stress over these questions to avoid their cardinal sin. Lost Sales. With competition constantly keeping brands on their toes to retain market share, companies definitely would not want to lose the sales because the supply chain could not make the product available for paying customers

6. Overall system performance – Big Data applications can encompass one’s end to end supply chain management and can enable the supply chain function to become nimble and flexible. Increased adoption allows managers to remain on top of all aspects of the performance through information providing as well as recommendations and predictions, quick slicing/dicing of data, as well as insights that are pushed to the function bearers for timely decisions and actions.

Having now understood the criticality of utilizing big data analytics within one’s supply chain, our next question would be what solutions can be used to consume and process big data efficiently. Big data solutions primarily deal with the challenges of volume (dataset size), Variety (Structures or types of data) and Velocity (speed or rate of transactions involved). The three types of big data analytics are –

Descriptive analytics – This could be looked at as an entry level into analytics where the aggregated big data is processed to glean relevant insights based on historical performance.

Predictive analytics – This is the next level where machine learning algorithms, as well as multiple statistical models, are applied to the historical data. This method can help users to predict the future outcomes

Prescriptive analytics – This is cutting edge analytics which employs a complex recipe of machine learning algorithms, business rules, computational modeling, and artificial intelligence developed on the basis of historical user feedback/activity to prescribe the optimal action for any pre-specified outcome.


Some of the commonly used tools for big data analytics are:

  1. Microsoft Excel: As most people are comfortable with Excel, there is a provision to connect the data stored in Hadoop (an open source java based big data storage framework), and to use the power view feature of excel 2013 for visualizations. This is normally used by small businesses as it isn’t a very scalable or robust tool to analyze or visualize your big data.

2. Business Intelligence Tools: BI tools mainly perform descriptive analytics, i.e., illustrate your data after it has been processed on storage platforms such as Hadoop, SQL, SAS, etc. for the ease of business managers. Some BI tools even allow you to connect with your SAP or Facebook account. BI tools have pre-built analytics as well as allow you to design your queries.

3. Machine learning based tools: These tools can offer predictive forecasting by leveraging advancements such as neural networks, and deep learning algorithms can discover hidden patterns in unstructured data sets and uncover new information. A neural network is a function that learns the expected output for a given input from training datasets. They are adaptive and able to modify themselves as they learn from subsequent inputs

4. AI-based tools: AI can help businesses to make intelligent decisions based on past business performance by learning from past data and intelligently forming relationships amongst various structured and unstructured data variables. Artificial intelligence tools can even implement business decisions autonomously with no human intervention. Google Maps utilizes AI to not only guide us to our destinations but also has the intelligence to suggest alternative routes which can take us to our targets faster. Walmart uses AI to analyse what customers are buying and when, how the product is trending on Twitter, how the weather might affect sales, and so on. Businesses are keenly following updates in Artificial Intelligence and Big Data Analytics, as these are among the most promising technology paths that businesses may take in the future. Gartner predicts that organizations will soon be competing largely on their analytics capabilities. It is expected that less than 5 percent of companies will be evolved to the sophistication level where they have developed and deployed artificial intelligence based on prescriptive analytics methods on their business. Embracing one’s big data will be a vital step in one’s journey towards being among the world’s crème de la crème companies.


Author Bio – Having 11 years of work experience, I have gained insights into various customer challenges and have offered holistic solutions which are tailored for the business. With a data-driven approach, I try and assist organizations in designing end-to-end supply chain solutions, optimizing organizational costs and productivity, and implementing innovative technology-based solutions. Being a writer by choice, I like to contribute to various quality websites especially those who talk about my area of interests. We grow by making people happy and successful.

Vivek M Kamath

Supply Chain & IT Advisory