Supply Chain Leaders Know They Have a Big Data Problem

by   |   August 8, 2012 4:19 pm   |   0 Comments

Today’s supply chains are more complex than before. While the structured data and the systems that use them will not go away, new forms of data offer new opportunities for companies to solve previously unanswered problems. These new data types—from mapping and GPS sensors, to voice, images and video—do not fit into traditional applications or data models. That’s the bad news.

The good news, as we learned in a survey of 53 IT and supply chain managers, is that companies are beginning to recognize that they have a problem and that they need to respond. While there is a general lack of understanding of big data terms and technologies, there is an awareness that supply chain best practices are moving from insights into supplies to leveraging insights into demand.

The survey found that:

  • Big data projects are underway. Today, 36% of organizations currently have a cross-functional team evaluating the potential of big data for their supply chains.
  • CIOs are in charge. The leader of the team evaluating big data usage and technologies for supply chain management is usually the Chief Information Officer (47 percent of respondents). The evaluation is less frequently led by a line-of-business leader (21 percent), or a cross-functional management team (21 percent).
  • System complexity is high. The average supply chain leader has four instances of enterprise resource planning (ERP) systems. The companies surveyed had 150 unique systems supporting their supply chains. See Figure 1 below:

Figure 1: Current System Complexity in Supply Chain Operations

Figure 1: Current System Complexity in Supply Chain Operations

  • Data is growing in the enterprise. Today, 8 percent of respondents have one petabyte of data in a single database, while 47 percent of respondents expect to have one-petabyte database within the next five years. Of those with big data initiatives underway, 68 percent expects to have a one-petabyte database within five years.
  • Variety of data types leads to variety of responses. Traditional supply chains have largely focused on transactional data, so it should come as no surprise that respondents who are evaluating big data initiatives have greater comfort with structured data types. The sample size is small—between 17 and 19 respondents answered each question in the Figure 2 below—but shows they are least comfortable with social data.

Figure 2: Companies Self-Rated Capabilities to Use Different Types of Data

Figure 2: Companies Self-Rates Capabilities to Use Different Types of Data

  • Current focus is on supply, but future sights set on demand data. The greatest perceived benefits— and the lowest current performance ratings—are in the area of demand data. However, due to the greater familiarity with transactional data and supply systems, respondents said the most common big data initiative focuses on supply chain visibility. This was frequently seen in retailers with long supply chains crossing many borders.

Figure 3 below summaries the survey responses views on two levels: what they view as the most important big data-related project, and second, how they evaluate their organization’s ability to use the various data sources cited. Again the responses reflect 19 companies that are have a big data initiative underway.

Figure 3: Importance and Current Performance of Big Data Projects in Companies with Big Data Initiatives

Figure 3: Importance and Current Performance of Big Data Projects in Companies with Big Data Initiatives


Getting Started
Clearly there is a lot of work to do for supply chain leaders who are already dealing with complex systems and now want to access new data types. Starting on this path means taking an inventory of all supply chain projects and evaluating whether each one is approaching the limits of traditional packaged applications. The ones that are require more advanced techniques.

Other initiatives to consider: look for data services that give your business insight into buy- and sell-side markets, such as Bazaarvoice for ratings and review data and Dun & Bradstreet for supplier data. Develop in-house sentiment analysis capability, such as through text mining applications, to be sure you can listen for the questions from customers that you don’t know to ask. Examine how data repositories from companies like Greenplum, IBM Netezza or Teradata can help you clean and prepare supplier and channel data for further analysis.  And make the use of analytics repeatable and systematic, not the specialty of an individual data scientist.

These efforts are not destinations by themselves, but steps on a longer journey. In a future column, I will offer recommendations for supply chain leaders who are looking to improve their business decisions-making—and profitability—by taking advantage of the proliferation of data types.

Lora Cecere

Lora Cecere, the founder of the Supply Chain Insights research firm, is the co-author, with Charles W. Chase Jr., of the forthcoming book Bricks Matter: The Role of Supply Chains in Building Market-Driven Differentiation (Wiley, 2012). You can read her blog, Supply Chain Shaman and follow her on Twitter @lcecere.

Tags: ,

Post a Comment

Your email is never published nor shared. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>