Available online at www.sciencedirect.

com

ScienceDirect
Materials Today: Proceedings 4 (2017) 1106–1115 www.materialstoday.com/proceedings

5th International Conference of Materials Processing and Characterization (ICMPC 2016)

Application of Big Data in Supply Chain Management

Ankit Singh*, Deepak Jain, Ishant Mehta, Jishnu Mitra, Saurabh Agrawal
Mechanical and Automotive Engineering Department, Delhi Technological University, Delhi-110042, India

Abstract

Big data is a term used to describe the massive volume of both structured and unstructured data that is too large that it is difficult
to process using conventional database and software techniques. Even though Big Data is in its early stages, it has remarkably
transformed nearly each and every sector in the industry. It has also revamped the Supply Chain Management giving it a new
dimension by increasing efficiency of production and optimization of operations. This paper identifies the issues regarding
Supply Chain Management by employing Delphi technique and aims to resolve them by incorporating Big Data
Analytics.Finally, an example of big data analytics application is also presented as a way of unveiling some of the relatively
unexplored territories in big data analytics research.
©2017 Elsevier Ltd. All rights reserved.
Selection and peer-review under responsibility of Conference Committee Members of 5th International Conference of Materials
Processing and Characterization (ICMPC 2016).

Keywords:Supply Chain Management; Delphi Technique; Big Data; Issues; Case Study

1. Introduction

1.1. Supply Chain Management

Supply chain management is the management of the flow of goods and services. It includes the movement and
storage of raw materials, work-in-process inventory, and finished goods from point of origin to point of
consumption.[1] Even though the definition might sound facile, it is quite a complicated process with multifarious
bottlenecks at different stages. Businesses from manufacturers, wholesalers and retailers, to warehouses, healthcare
providers and government agencies use supply chain management principles to plan, assemble, store, ship, and track
products from the beginning to the end of the supply chain.
*
Corresponding author. Tel.: +91 9953159447
Email address:ankit10sevs@gmail.com

2214-7853©2017 Elsevier Ltd. All rights reserved.

Selection and peer-review under responsibility of Conference Committee Members of 5th International Conference of Materials Processing and
Characterization (ICMPC 2016).
 Ankit Singh et.al. /Materials Today: Proceedings 4 (2017) 1106–1115 1107

Fig. 1. Supply Chain Model[2]

Figure 1 represents the model of supply chain management .The main role of a supply chain in an industry is
discussed in as shown in below table 1:

Table1 . Role of Supply Chain Management in an industry

SUPPLY CHAIN MANAGEMENT

PLANNING
1 Deamnd Forecasting
2 Inventory Control
3 Customer Service
4 Purchasing
1.2. Big Data
OPERATIONS
1 Day to Day management
2 Storage
3 Distribution
4 Fulfillment

Big data is a term used to describe the exponential growth and availability of data, both structured and
unstructured and big data may be as important to business – and society – as the Internet has become.

1.2.1. The Three V’s of Big Data

Table 2: Causes of Big Data[3]
Type of Data Volume
More detail around the sale,
including price, quantity, items
Sales sold, time of day, date, and
customer data.
Velocity Variety
From monthly and weekly to Direct sales, sales of distributors,
daily and hourly. Internet sales, international sales,
and competitor sales

More detail regarding decision From click through to card usage. Face profiling data for shopper
andpurchasing behavior, identification and emotion
includingitemsbrowsed and detection; eye-tracking data;
Consumer bought, frequency,dollar value, customer sentiment about
and timing. products purchased based on
“Likes,” and product reviews.

Perpetual inventory at more From monthly updates to hourly Inventory in warehouses, stores,
locations, at a more disaggregate updates. Internet stores, and a wide variety
Inventory level (e.g., style/color/size). of vendors online.

Location and Time
Sensor data to detect location in
store, including misplaced
inventory, in distribution center
(picking, racks, staging, etc.), in
transportation unit.
Frequent updates for new
location and movement.
Not only where it is, but hat is
close to it, who moved it, its path
to get there, and its predicted path
forward; location positions that
are time stamped from mobile
devices.

Currently the world has around 2.8 zeta bytes of data and by 2016 the cumulative size of the entire world’s data
is expected to exceed 16,000 acres and by 2020 this data is going to become 50 times larger and the same is depicted
1108 Ankit Singh et.al./ Materials Today: Proceedings 4 (2017) 1106–1115

within the figure 2 a and 2 b. It is estimated that only 33% of this information can be appropriately targeted and
analyzed but currently only 0.5% of this information is only analyzed as shown in above table 2. [4]

Fig. 2. (a) Data Distribution in 2012; (b) Expected Data Distribution in 2020[4]

2. Literature Review and Research Gap

Supply Chain Management is a way of managing the business and its relationships.[5]The conventional meaning
of supply chain management needs to be transformed with the upcoming trends in SCM. With the dramatic changes
taking place within a span of few years in this industry, the SCM needs to be adequately updated and adapt
accordingly to the needs of the industry.
With the introduction of Reverse supply chain which is defined as the process of planning, implementing and
controlling the backward flows of raw materials, in process inventory, packaging and finished goods, from a
manufacturing, distribution or use point to a point of recovery or point of disposal [6], which has made the supply
chain complex and a dynamic value web hence making the optimization of operations intractable.Hence, the older
techniques need to be updated and new technological changes are the need of the hour.
We concur with Waller and Fawcett who more or less argue that previous research have not yet properly closed
the gap between supply chain functional knowledge. The research papers provide clarification on key aspects of
SCM that will aid practitioners and researchers in their desire to understand and implement SCM,[5] but sometimes
find it more difficult to see the direct connection of big data to SCM.[3] Even though these research papers have
found solutions to the existing problems in SCM, they find it difficult to concentrate on each aspect and optimize
every function.
Chaeexamined the role of social media platforms in generating big data useful within supply chains. [8]Ivanov
and Sokolov provided an inter-disciplinary modeling framework for model-based decision support within
collaborative supply chains.[9] Sensors and transaction logs within organizations, as well as Internet linkages both
from fellow supply chain participants as well as from other Internet sources have led to an explosive growth in big
data, which can be valuable to track materiel by location, and can be useful for supply chain planning.[10]Taguchi
method used by Shouvik and Prasantais a powerful tool for designing high quality systems with an optimum setting
of process control parametersintroduces an integrated approach that is simple and efficient to find the best range of
designs for quality, performance and computational cost in SCM. [18]
The emergence of new terms, such as SCA, reflects a broad interest in leveraging the business value of supply
chain data and harnessing the power of various analytical technologies and methods. Top performing companies are
better at utilizing their data for business planning and execution and this has led to the increase in supply chain
integration and visibility. In general, academic research expects the benefits of analytics in supporting supply chain
operations.[11]
Thus after reviewing studies and research papers on supply chain management, we have identified the research
gap in supply chain management which requires technological changes along with better implementation to tackle
the upcoming and the current issues subsequently. Thus we need to come up with the software that provides better
 Ankit Singh et.al. /Materials Today: Proceedings 4 (2017) 1106–1115 1109

data analytics along with easy implementation to help sought the issues pertaining to SCM. Big Data Analytics can
be one of the best solutions to the SCM issues, which can not only store and utilize the data successfully, but even
provide the best solutions.
Only 20% of the companies have taken up Big Data Analytics. This implies that there is much to be understood
and worked upon in Big Data analytics’ implementation in SCM. This gives us further prospects of working and
successfully incorporating the big data techniques into SCM.

3. Research Objective

Fig.3. Supply Chain Evolve into value web[5]

As can be seen from the above figure 3earlier the supply chain had a linear relationship between the suppliers,
manufacturers, distributors and the consumers but now with the advent of e-commerce and rising customer demands
the supply chain has become complex and hence the managers are finding it difficult to optimize the operations.
So the main objective of the research is to find a way to reduce the complexity of the supply chain by using Big
Data Analytics and to identify the data sources that may be considered for decision-making purposes in each of the
SCM levers: procurement, marketing, transportation, and warehouse operations.
Also the research is aimed to identify the issues of supply chain management that can be resolved using Big Data
Analytics and to find solutions to them by applying Delphi Technique.

4. Research Methodology

The research methodology delineates the processes and methods equipped to conduct the research on any subject. It
includes basically three steps: collecting material, descriptive analysis and final evaluation.

One of the most effective research methods that have been used recently is the Delphi Technique. The Delphi
Technique is an effective tool for thorough research and analysis of any issue. The Delphi Technique allows
individuals to get in depth knowledge of any topic by asking the experts. The experts are able to guide the
individuals on the issues providing them with personal experiences and proper guidance.
Online websites like OpenIDEO, Techcast and Horizon serve as medium to these techniques. It allows people to
interact with individuals personally online. There is usually a panel of 3-4 individuals, each expert in their own
domain. We have utilized the Delphi Technique to get acclimatized to the issues of supply chain management and
the tree diagram for the application of Delphi technique is shown in figure 4. The experts have delineated the issues
in recent times and the various techniques that were being used to solve the upcoming issues. Figure 5 shown below
representation of aquestionnaire which was presented to the experts to get insights on the issues and their solutions,
1110 Ankit Singh et.al./ Materials Today: Proceedings 4 (2017) 1106–1115

Fig. 4 . Application of Delphi Technique

Fig. 5 . Questionnaire

The questions were answered by experts that have helped us in finding the solutions to these issues that can be
successfully resolved using Big Data.
The other method that has been utilized is the study of different case studies. The inclusion of the case study is
done to maintain practicality at the core. Case Studies investigated simultaneous big data analytics examples,
typically in emerging practices, thus being a successful way of including the latest trends detected in the industry.
[8].We have traced the case study of United Parcel Service (UPS)to gain practical experience of what can be
achieved through incorporation of Big Data Analytics in SCM.
 Ankit Singh et.al. /Materials Today: Proceedings 4 (2017) 1106–1115 1111

5. Research Findings

One of the most significant paradigm shifts of modern business management is that individual businesses no
longer compete as solely autonomous entities, but rather as supply chains. Business management has entered the era
of internetwork competition. Instead of brand versus brand or store versus store, it is now suppliers - brand - store
versus suppliers – brand - store, or supply chain versus supply chain. [5]
Any company that has undertaken the mission of implementing a supply chain management strategy knows that
one of the greatest challenges it faces is the significant change in internal culture that is required to make the supply
chain redesign successful. However difficult it may be to accomplish, change can be implemented successfully
when directed by a strong and knowledgeable leader, who understands the tools available for achieving positive
change, as well as their role in initiating and sustaining these changes.

5.1 Issues in Supply Chain management

5.1.1. Primary Issues

• Reducing Operating Cost: One of the most common goals for any industry includes cost reduction and is so
applicable for Supply Chain Management as well.
• Overall Inventory Cost (Shorter Lead Time and Better Throughput): With shorter product life cycles
andchanging market demands, companies are forced to embark on a lean journey. Realistic forecasting of
future sales is the need of the hour for optimizing the inventory and reducing its cost. [13]

5.1.2. Secondary Issues

• Improve Customer Service: A number of industries have established themselves in the Supply chain
industry and providing customers with more and more services making it compulsory for other industries to
provide the same or even furthermore to stay in the market and compete with their rivals.

• Speed of Product Delivery to Markets: In recent times, with quick and timely deliveries from various
industries has made it mandatory for other companies to match up their delivery time. It has become
essentially a requirement for the companies and natural for the customers to have their goods delivered
within a limited period of time.

• Visibility: With the ever growing competition in the supply chain management and new companies
foraying their way in the supply chain industry, the products are differentiated and must match the
customer requirements.

• Risk Management: Companies that are too reliant on one supplier are vulnerable if that supplier can’t meet
demands. This means that supply chain managers must mitigate risks, for example an earthquake in Asia
should not halt the supply in North America. Both mitigation and contingency strategies must be taken care
of by an industry to ensure minimum disruptions in supply due to external and natural factors.[14]

• Increasing Customer Demands (Growing Expectations): With in numerous ways of advertising products to
each household, makes it further more competitive for each firm to not only better display these products
but provide the best product offered in the market. Also each company needs to keep itself updated with the
market, and develop subsequent strategies to tackle other companies in ever growing cut throat competitive
market. [15]

• Globalization: With globalization, each and every product that is being sold must match the quality of
products made in other countries for customer satisfaction. Today people are well aware of the products
being offered in the outside world, and hence are intelligent enough to distinguish between good and just
satisfactory products. The companies are hence required to provide the best they have to their customers. In
1112 Ankit Singh et.al./ Materials Today: Proceedings 4 (2017) 1106–1115

these cases a company is required to have global procurement network which can support and react to their
supply chain needs. Also a company needs to select suppliers that provide manufacturing locations with
consistent global quality and a reliable local service to compete effectively in the market.[16]

• Making Supply Chain Less Complex: With so many suppliers, manufacturing location, different products
and varied costumer needs make the supply chain complex. Thus making supply chain less complex is also
required for increased efficiency and optimized
operations.

Fig. 6 . Issues in Supply Chain[12]

6. Solutions to issues using Big Data

• In Reducing Delivery Time

The organizations which adopt big data analysis will have the capacity to connect their systems with
systems of millions of their customers. This would allow them to move the data ahead of packages. For
example, for a delivery company, packages arriving from a different city could be directly send to a
warehouse near the intended delivery address and shipped out to their destination the next day without any
delay. This would save valuable time and effort which is wasted in storing all the packages arriving at a
particular city at the same warehouse, and then sorting them according to localities.
Big data analytics would also enables companies to develop algorithms that can predict the maintenance
requirements of their systems. With the multitude of products produced and delivered daily, an unplanned
breakdown of a system can have serious consequences resulting in late packages and angry customers.
Analyzing the data from the various machines involved in production, or analyzing the data from the
delivery vehicles, would give a signal as to when should the machine or vehicle be serviced. This would
give the organization time to arrange for a back-up mechanism. This would help in keeping the delivery
time unaffected. Predictive maintenance would save millions in preventing unnecessary and unplanned
maintenance and would also keep the customers pacified.
• In Improving Customer Service
To keep a customer loyal to the organization, the customer needs to be pampered with newer services
and offers frequently. To know which offer or service is apt for a particular customer, the organization need
to know the likings of the customer. This is where big data comes in. Collecting data about the buying
habits, delivery choices, payment methods, shopping seasons etc of the customer would give the
organization proper knowledge about the consumer. Using this knowledge, the apt offers and services
could be pitched to the customer. This would make the customer feel like he or she is receiving personal
care. And pitching the right offers and services, would also lead to increase in sales.An example of using
big data to provide improved customer care services could be found in the delivery business. Data could be
collected regarding when a customer is available to collect the package, based on previous experiences.
This would allow the company to decide the windows of time, when the packages should be delivered to
 Ankit Singh et.al. /Materials Today: Proceedings 4 (2017) 1106–1115 1113

respective customers. Receiving packages at their preferred time slots would give a positive view of the
company to the customers, and encourage them to choose the company in the future as well.

• In Reducing Inventory Costs

The purpose of inventory is to act as a buffer in cases pertaining to sudden increase in demand, sudden
shortage of supply of essential raw materials and parts from the supplier or break down of machinery.
Using big data can drastically reduce the level of inventory needed to be maintained. Data related to past
demand can be used to accurately predict future demands. Similarly, the history of the supply patterns of a
particular supplier can be used to predict the situations in which a shortage of supply can be expected. And
analyzing performance data of the machines can help in formulating preventive algorithms to service the
machines before they break down. Using these techniques can significantly reduce the inventory levels,
leading to savings of wealth and space.

• In Reducing Operational Cost

Reducing operational costs is a primary area of concern for most industry leaders. Reduced operational
costs lead to increased profits and greater productivity. But operational cost is a function of costs associated
with multiple variables like raw materials, maintenance of machineries, outsourcing, inventory, electricity,
transportation and many more. All these variables are interlinked. An alteration in any one of them might
have an adverse effect on the others. For example, a reduction in inventory of raw materials will lead to
reduction in inventory costs. But if the reduction is too much, then the extra costs incurred due to the
increased frequency of raw material transportation will outweigh the reduction in inventory costs. Hence
the reduction of overall operational costs is a complex problem. The primordial task is the selection of
variables, which can be altered or modified to the right extent, to get substantial reductions in their costs,
without increasing other costs. If a firm tries to tackle the majority of the variables simultaneously, without
creating a priority order, then in invariably leads to wastage of resources, without getting any appreciable
results.
Big Data Analysis can play a decisive role in this process of prioritization. It can also play a role in
balancing the reductions in each variable cost so that they do not increase other costs. Considerable volume
of data relating to each of the variables can be collected. Analysis of these data, would give the firm
insights into the independent cost functions as well as the relationship between different variables. This
would enable the firms to prioritize variables for alterations. Using these analyses, precisely calculated
changes (based on the analyses) can be made to prioritized individual variables, which though affecting
other variables, would lead to overall reduction in operational cost. Different models can be created using
different alterations based on the analyses.

7. Case Study

7.1 United Parcel Service

United Parcel Service (UPS) is the biggest package shipping company in the world. It was founded in 1907 and
on a daily basis they make more than 16 million shipments to over 8.8 million customers globally. In addition, they
receive on average 39.5 million tracking requests from customers per day. They gather data at every possible
moment and already store over 16 petabytes of data.
They use Big Data to analyze such huge data. UPS uses proprietary package flow technology to determine what
packages are loaded on each vehicle, then gathers data from several aspects of fleet operations using a telematics
technology system.
UPS uses telepathic trucks which have sensors that capture over 200 data points for more than 80,000 vehicles
every day. The data analyzed includes:
• Engine Monitoring
• Speed
• Mileage
1114 Ankit Singh et.al./ Materials Today: Proceedings 4 (2017) 1106–1115

• Number of Stops
• Miles per gallon
• Safety aspects

Analyses on these parameters have resulted in exponential results:

• 2.6 million X 10 million minutes of idling time reduced in 2012, saving more than 1.5million gallons of
fuel.
• 13000 metric tons of carbon emissions avoided due to reduced miles per stop- 12.1 million miles of driving
eliminated in 2012.
• 39 million x 1 million gallons of fuel saved since 2001 due to route optimization.
UPS takes a holistic approach to integrating data into technology. Arguably the world's largest operations
research project, ORION uses fleet telematics and advanced algorithms to take route optimization to a new level. In
2013, UPS began the first major deployment of ORION, with plans to deploy the technology to all 55,000 North
American routes by 2017. [17]

8. Conclusion

From our study on Supply chain management, we can infer that the structure of supply chain has transformed
rapidly in the past few years. Thus to overcome this complexity in supply chain and handle the humongous data that
has compiled over the years, we need a powerful tool. Big Data provides us with the perfect solution, by not only
abating the complexity, but also successfully optimizing the operations.
The above stated results along with the successful story of the above company, provides enough evidence and
supports the big data implementation in the supply chain industry.
Thus we can see that big data analytics can be successfully implemented in Supply chain management producing
fruitful results for the organization. This integration has allowed companies to come up with highly innovative
solutions and accelerate the widespread adoption of these core supply chain principles.
Also we intend to work further on big data analytics using hypothetical data to further resolve the remaining
issues of supply chain.

References

[1]OBE, Graham Hoyle, Association of Employment and Learning Providers, January 3, 2013.

[2] Akcali E., Etinkaya S. and Uster H. , Networks, 2009: 231-248.

[3]Fawcett, Matthew A. Waller and Stanley E. , Journal of Business Logistics, 2013: 77-84.

[4]Davenport, Thomas H. , International Institute of Analytics, SAS, 2013.

[5]Douglas M. Lambert, Martha C. Cooper. , Industrial Marketing Management. New York: Elsevier Science Inc, 2000. 65-83.

[6]Navendu Sharma, Amish, ManishaMeena,Saurabh Agrawal. , 4th International Conference on Materials Processing and Characterization.
Hyderabad: Elsevier, 2015. 3100-3016.

[7] Marchese, Eamonn Kelly and Kelly. , Business ecosystems come of age report. Texas: Deloitte University Press, 2015.

[8] Chae, International Journal of Production Economics, 2015: 247-259.

[9] Ivanov, D., and Sokolov, B. , Journal of Manufacturing Technology Management, 2012: 976-997.

[10]Olson, David L. , Lincoln: College of Business Administration, 2015.

[11]Bongsug (Kevin) Chae, David Olson &ChwenSheu. , International Journal of Production Research, 2013.

[12]Braunstein, Cal. , Wilton: Robert Frances Group, 2013.

 Ankit Singh et.al. /Materials Today: Proceedings 4 (2017) 1106–1115 1115

[13] Dean, David. , Zug: Bossard Group, 2012.

[14]Montroy, Jeannette. , Canada: Procurement Sense, 2015.

[15] O’Marah, Kevin. , Supply management, October 1, 2013.

[16] Berrios, David, Madison: Wisconsin School Of Business, 2014.

[17]Jack Levis, Dale Spencer., California: Mother Nature Network, 2013.

[18] Shouvik Ghosh, PrasantaSahoo, GoutamSutradhar., International Journal of Advanced Materials Manufacturing and Characterization.
Hyderabad: Advanced Materials Manufacturing & Characterization, 2014. 93-99.