Supply Chain Analytics: The New Frontier

Posted in Operations & IT Articles, Total Reads: 2993 , Published on 09 January 2013

Supply chains cannot tolerate even 24 hours of disruption. So if you lose your place in the supply chain because of wild behavior you could lose a lot. It would be like pouring cement down one of your oil wells.  - Thomas Friedman

Supply Chain is the most important component of any organization. Nothing can function effectively without the delivery of its raw materials. Imagine a plant functioning without raw materials, or a football match happening without a football. From our daily routine to complex processes that are carried out in the Industry, supply chain management is inherent. It is defined as joining of all the processes from the suppliers to the end consumers, adding value at each phase and hence delivering a high quality product to the end user.

In today’s world with cut-throat competition all the companies are striving to attain excellence in their respective fields, The most critical being having an efficient supply chain network. It has led them to take calculated risks. Here the word “Calculated” is the key. Gone are the days when decisions were based on the intuition and gut feeling of the manager. Inventory management, Raw materials purchase, WIP (Work in Progress Management), Lead time & logistics are maintained by taking the assistance of software which helps in the efficient working of the Organization. SAS, the world’s leading business analytics software Services Company’s CMO lately came up that Supply Chain Analytics comprises of 8 levels:



Standard Reports

Generated on a regular basis. Ex monthly reviews, financial reports

Ad Hoc Reports

Allows an individual to ask a few questions related to the issue


Query bases questions, allows a certain level of research


Notifies when something goes wrong to prevent it happening

Statistical Analysis

To Analyze a particular issue


Mostly used for estimation, provides the framework for the future

Predictive Modeling

Predicting the segment & making future analysis for target segment


Best way to solve the issue, selection from multiple choices

Earlier, when Industries were either government controlled or Monopolistic, Supply Chain Management (SCM) was taken for granted. People would wait for years to get a scooter or a telephone. The paradox was in spite of being the backbone of an organization; it was never focused upon due to lack of competition. After the entry of private players, organizations realized the importance of having an efficient supply chain installed in their system. But the problem stood tall: How is the level of efficiency of the supply chain measured? How the high level of standards are maintained or achieved? This is where Supply Chain Analytics came into picture. Supply Chain Analytics provides the proper framework for effective functioning of the Industry. Employing the proper techniques to manage supply chain results in better lead times, lower Inventory holding costs, maintaining optimum levels of WIP, raw materials & finished goods. The benefits are endless. It finds immense usage in the four phases of SCM i.e. Plan, Source, Make & Deliver. In Planning, proper analysis is done so as to understand the market demands or needs, changing trends etc. In Source, historical data is employed in analyzing the suppliers etc which directly affect costs. In Make, production/batch size, storage time etc is taken care of. In Deliver, areas more importantly related to Logistics/dispatch/COMD (Customer Order Management Department) are taken care of. On a broad scale, Supply Chain Analytics can be divided into three parts as shown in the figure.

1. Purchase Analytics: Purchase is one of the first departments that come into picture in the functioning of the organization. It is this department which decides the purchase of Raw materials, parts & spares etc. Inefficient purchase can cause a setback for the company even before it starts manufacturing. It is imperative that the company purchases the required goods at proper prices. It is also required to ensure the credibility of the supplier so as to maintain that the inflow of raw materials is not delayed. It also helps in maintaining Inventory level times over vendor/supplier, maximizing profit margins on goods purchased yet maintaining quality is also essential.

To look into the above-mentioned concerns, companies have come up with their own set of purchase Analytics tools in order to keep a check on their KPIs (Key Performance Indicators). Companies like APTEAN, Profitbase & CHAINalytics etc provide their services to companies based on the type of business they are involved in. It includes a number of activities. Some advantages are mentioned below.

a. Adequate Business Insight: This involves the usage of proper Purchase Analytics that enable the company to understand issues like timely deliveries, material cost reduction etc

b. Vendor Ranking/Rating: The suppliers are ranked based on the quality of product/ raw material delivered. They are also ranked on the level of timely deliverance. This helps the organization to get a better insight on the reliability of the vendor. This also acts as a tool which provides a mutual benefit. Vendor Score cards provide a quantitative estimation of the reliability, quality & quantity of product delivered. This generates competition and in turn helps the suppliers to improve their processes so as to maximize their efficiency.  A snapshot of the Vendor Score Card is shown below.


The above snapshots show how suppliers are rated on different parameters. It also helps organizations in choosing their supplier based on the criticality of the project. For example, a project does not require immediate delivery but costs are more important, in that case, the most economical supplier is selected. Another project requires immediate delivery with high profit margins; the organization could go for a supplier reliable in terms of delivery.

2. Warehouse Analytics:  Warehousing Costs turn out to be one of the heftiest bills on a company. In order to have a low lead time, companies produce non-customized goods so as to keep the stock ready when the demand comes. For tailor-made industries, the raw materials and WIP are stored and kept for further processing when an order is received. But the biggest question that baffles plant managers is: How to forecast the demand? Also, what is the safe level of Inventory that should be kept so as to deliver the finished goods as fast as possible and also minimize storage cost.  Warehousing cost at times turn out to be as high as 30% of the product value. This in turn becomes a big issue for companies operation on a thin margin. To forecast the optimum Inventory Stock, proper analytic tools are required to balance the costs and enable the company to make a decent profit.

The figure below shows the overall process of Data-warehousing Analytics that is used in SCM. The historical data is used and loaded into the end segment using ETL (Extract, Transform and Load) and gives us an optimum result through tools like data mining etc. the output hence received through data mining can also lead to wrong results or totally disconnected results but by the usage of proper inputs, it is still a significantly reliable way of converting historical data (provided by Pre-Data Warehouse)   into meaningful results that can benefit the organization.

Overview of Data Warehouse Analytics



The measurement of effectiveness is evident through several KPIs (Key Performance Indicators). The above process enables analyzing and forecasting the optimum shipping mix based on the criticality of the dispatch. It helps the manager take multiple operational decisions with greater confidence. It also keeps a check on the delivery rate based on the time taken to fulfill the same. The Analytics also ensure maximum utilization of the resources available. Optimum level of Inventory is managed by forecasting the demand which drastically reduces holding & maintenance costs. All of this in turn leads to customer delight which improves the image of the organization in terms of its ability to deliver the goods in time.

3. Sales Analytics:  Sales determine the bottom line of a company. The amount of revenue an organization will generate directly depends on the sales output. Hence, it is imperative that a company is able to determine how it is doing, how its sales people are performing, If the customer is satisfied with the product etc.  All the organizations focus on an accurate forecasting. According to Aberdeen Group, a research group has emphasized the importance of sales analytics. It says that the companies using sales analytics has outperformed companies which underestimated the importance of the same. Forecasting using sales analytics has been considered as the tool which shows the “Big Picture”. Sales Analytics also answers the typical questions as prioritizing their products based on demand. It also helps in understanding the threats and investing on opportunities. Sales analytics gives the manager the understanding of fragmented demand and how to understand user’s demands and focus accordingly.



For example, as shown in the above figure is the software (xceler8) provided by Crown, a U.S. based sales analytics company. It shows how the user (manager) can see the regional sales of the product. Such flexibility allows an individual to focus on the markets and thereby control the supply of finished products into that zone depending on the demand. It allows companies to select their target market based on age, income levels etc. Organizations understand that the buyer is not always the consumer. They are also aware of the fact that the group with the highest disposable income may not be their most potential buyers. Companies are interested in knowing the breakup of the sales in a given area or zone. This helps them to understand the taste and preferences of a particular good.


The figure shown is the snapshot of sales analytics software omnibus 4.0; it shows the sales breakup of the different types of shoes. In addition to that, it also provides a graph which serves as the basis for comparison of performance.

Supply Chain Analytics: Limitations

Despite several breakthroughs in Data Analytics, the individual has to analyze huge amounts of data. Data mining at times leads to totally disconnected results. For example, improper selection of inputs might lead to a doubtable outcome like “the usage of Colgate toothpaste increases during the wars.” This stands as a major challenge in the implementation of Data Analytics as there are several thousand ways to analyze it. To overcome such hurdles, Dr. Micheal Watson, a known figure in Supply chain analytics has broken the problem into three simple and easy to understand parts. Knowledge of these three sections can help an individual in analyzing data efficiently & effectively.

Figure: the different parts of Analytics to enable understanding

  1. Descriptive Analysis: This is used to better understand one’s own system of supply chain. Historical data provides as the background of all the research that is carried on it. For example: Checking past shipment records might help one understand its current shipment timing.
  2. Predictive Analysis: This is used mostly to forecast the future demands by analyzing historical or past data. For example, a company evaluates its sales of tin cans in the winter by analyzing the demand & sales patterns of the winter sales of last 5 years.
  3. Prescriptive Analysis: this type of analysis is used when a solution or value is required. For example, a company sets the inventory level for its tailor made goods by analyzing the previous patterns.



Future Scope

With more and more companies realizing the power of Supply Chain Analytics, it is almost certain that its growth trajectory will only know an upward trend. It is also supposed that such easy to access technology will also promote sharing of knowledge and thereby eliminating “Knowledge Silos”, a problem very common in organizations with low manpower. Since whatever the supply chain analytics do rest on data & data doesn’t lie. It can be expected that organizations, both mid-sized and large would adopt these technologies based on their requirements.

This article has been authored by Mohit Kumar & Subhra Orhaw from IIFT.


If you are interested in writing articles for us, Submit Here