Beruflich Dokumente
Kultur Dokumente
Executive Introduction
Big Data and logistics are made for each other, and today the logistics industry is positioning itself to put
this wealth of information to better use. Big Data is a relatively untapped asset that companies can exploit
once they adopt a shift of mindset and apply the right drilling techniques. It also goes way beyond the
buzzwords to offer real-world use cases, revealing what is happening now, and what is likely to happen in
the future. This trend report starts with an introduction to the concept and meaning of Big Data, provides
examples drawn from many different industries, and then presents logistics use cases.
This is a frequent and disappointing experience Fig-7: Relative deviation of prognosis from
for shoppers: once they find the perfect item of actual sales volume, from: “Big Data &
clothing, they discover that the size they need is Predictive Analytics
out of stock. With increasing competition in the
textile and clothing segment, the availability of
popular garments is now typically limited. This is
caused by consolidation of brands and
accelerated product cycles. In some cases, there
are only three weeks between the first design of
a garment and its in-store arrival. The frequent
2.4 New Business Models driving business innovation outside its core
business units and brands. As part of Telefonica
Crowd analytics deliver retail and Digital, the “Dynamic Insights” initiative has
advertising insights commercialized the analysis of movement data,
creating incremental revenue from retail,
To provide effective mobile voice and data property, leisure, and media customers. Other
services, network operators must continuously carriers have developed similar offerings, such as
capture a set of data on every subscriber. Apart Verizon’s “Precision Market Insights” service.
from recording the usage of mobile services (for Fig-8: Analysis of customer footfall in a particular
accounting and billing purposes), operators must
also record each subscriber’s location so it can
direct calls and data streams to the cell tower to
which the subscriber’s handset is connected.
This is how every subscriber creates a digital trail
as they move around the provider network. And
in most countries it is just a small group of
network operators that have captured most of
the population as customers – their combined
digital trails of the subscriber base provide a
comprehensive reflection of society or, more
precisely, of how society moves. In the past,
organizations could only make internal use of
location and usage data from mobile networks. location based on mobile subscriber data
This is because of privacy laws that limit the
exploitation of individual subscriber information. 2.5 Data-Driven Supply Chains
But once subscriber identity has been split from
the movement data, substantial business value Few areas of business have been transformed by
remains in this anonymous crowd data, as big data analytics as much as supply chain
Telefonica has discovered. management. Same-day delivery has become
In urban areas, the density of digital trails is nearly mandatory to modern multichannel
sufficiently high to correlate the collective retailing. 38 As consumers, we have developed
behavior of the subscriber crowd with this expectation. We don’t think about it unless
characteristics of a particular location or area. there is a problem. It may be that the item we
For example, it’s possible to assess the ordered online doesn’t show up as scheduled or
attractiveness of a specific street for opening a an advertised item is out of stock when we try to
new store, based on high-resolution analysis of purchase it. Achieving a competitive level of
how people move and rest in this area, and find global supply chain excellence cannot be
the opening hours likely to create maximum accomplished without data-driven, end-to-end
footfall (see Figure 8). In a larger context, it is operations. Consider companies such as Tesco.
also possible to see the impact of events such as The company gathers huge amounts of customer
marketing campaigns and the opening of a data from its loyalty program. It then mines this
competitor store by analyzing any change in data to inform decisions from promotions to
movement patterns. When gender and age strategic segmentation of customers. Amazon
group splits are included in the data, and geo- came early to the frontier of data analytics. The
localized data sets and social network activity are online retailer pushed the frontier using
included, this segmentation adds even greater customer data to power its recommendation
value for retailers and advertisers. engine “you may also like...” based on a type of
With the launch of the Telefonica Digital global predictive modeling technique called
business division, the network operator is now collaborative filtering.
The company continues its rapid leadership in extended enterprise has provided transparency
fulfillment capabilities through data-driven and enabled coordinated cross-enterprise
decisions. Walmart was also an early adopter of efforts.
data-driven supply chains. By making supply- Big data analytics is the game changer. It has
and-demand signals visible between retail stores given rise to the intelligent supply chain.
and suppliers, the company optimizes all its
supply chain decisions—from customer
fulfillment to inventory tracking to automatic 3.0 Big Data in Logistics
purchase orders through its supplier portal.
The number of RFID tags sensing inventories Companies are learning to turn large-scale
across supply chains is in the millions. In fact, the quantities of data into competitive advantage.
number of RFID tags sold globally is projected to Their precise forecasting of market demand,
rise from 12 million in 2011 to 209 billion in 2021. radical customization of services, and entirely
Supply chains are increasingly combining data new business models demonstrate exploitation
from different systems to coordinate activities of their previously untapped data. As today’s
across the supply chain end-to-end. Marketing is best practices touch many vertical markets, it is
generating huge volumes of POS data from retail reasonable to predict that Big Data will also
stores that is automatically shared with suppliers become a disruptive trend in the logistics
for real-time, stock-level monitoring. RFID tags industry. However, the application of Big Data
monitor inventory on shelves and in-transit analytics is not immediately obvious in this
coordinating sector. The particularities of the logistics
with current stock levels for automatic order business must be thoroughly examined first in
replenishment. Add to this data from computer- order to discover valuable use cases.
aided design, computer-aided engineering,
Computer-aided manufacturing, collaborative
product development management, and digital 3.1 Logistics as a Data-driven
manufacturing, and connect it across
Business
organizational boundaries in an end-to-end
supply chain. Even more value can be unlocked A kick-start for discussion of how to apply Big
from big data when companies are able to Data is to look at creating and consuming
integrate data from other sources. This includes information. In the logistics industry, Big Data
data from retailers that goes well beyond sales. analytics can provide competitive advantage
It may be promotion data, such as items, prices, because of five distinct properties. These five
and sales. It also includes launch data, such as properties highlight where Big Data can be most
specific items to be listed and associated ramp- effectively applied in the logistics industry.
up and ramp-down plans. It also includes They provide a roadmap to the well of unique
inventory data, such as stock levels per information assets owned by every logistics
warehouse and sales per store. This data is provider. In the following sections, we identify
essential for the supply chain to deliver the items specific use cases that exploit the value of this
that are needed when they are needed. information, and contribute to operational
Through collaborative supply chain management efficiency, a better customer experience, or the
and planning, companies can mitigate the development of new business models.
bullwhip effect and better smooth out flow
through the supply chain. Many companies
guard customer data as proprietary, but there
are many examples of successful data sharing.
Walmart is a great example of requiring all
suppliers to use its Retail Link platform. The
exchange and sharing of data across the
3.2 Use Cases – Operational location information posted by recipients in
order to avoid unsuccessful delivery attempts.
Efficiency In summary, every delivery vehicle receives a
A straightforward way to apply Big Data continuous adaptation of the delivery sequence
analytics in a business environment is to that takes into account geographical factors,
increase the level of efficiency in operations. environmental factors, and recipient status.
This is simply what IT has always been doing – What makes this a Big Data problem? It requires
accelerating business processes but Big Data the execution of combinatorial optimization
analytics effectively opens the throttle. procedures fed from correlated streams of real-
time events to dynamically re-route vehicles on
the go. As a result, each driver receives instant
3.2.1 Last-mile optimization
driving direction updates from the onboard
A constraint in achieving high operational navigation system, guiding them to the next best
efficiency in a distribution network occurs at the point of delivery.
“last mile”. This final hop in a supply chain is
often the most expensive one. The optimization Crowd-based pick-up and delivery
of last-mile delivery to drive down product cost
The wisdom and capacity of a crowd of people
is therefore a promising application for Big Data
has become a strong lever for effectively solving
techniques. Two fundamental approaches make
business problems. Sourcing a workforce,
data analysis a powerful tool for increasing last-
funding a startup, or performing networked
mile efficiency. In a first and somewhat
research are just a few examples of
evolutionary step, a massive stream of
requisitioning resources from a crowd. Applied
information is processed to further maximize the
to a distribution network, a crowd-based
performance of a conventional delivery fleet.
approach may create substantial efficiency
This is mainly achieved by real-time optimization
enhancements on the last mile. The idea is
of delivery routes. The second, more disruptive
simple: Commuters, taxi drivers, or students can
approach utilizes data processing to control an
be paid to take over last mile delivery on the
entirely new last-mile delivery model. With this,
routes that they are traveling anyway. Scaling up
the raw capacity of a huge crowd of randomly
the number of these affiliates to a large crowd of
moving people replaces the effectiveness of a
occasional carriers effectively takes load off the
highly optimized workforce.
delivery fleet. Despite the fact that crowd-based
Real-time route optimization delivery has to be incentivized, it has potential to
The “traveling salesmen problem” was cut last-mile delivery costs, especially in rural
formulated around eighty years ago, but still and sparsely populated areas. On the downside,
defines the core challenge for last-mile delivery. a crowd-based approach also issues a vital
Route optimization on the last mile aims at challenge: The automated control of a huge
saving time in the delivery process. Rapid number of randomly moving delivery resources.
processing of real-time information supports This requires extensive data processing
this goal in multiple ways. When the delivery capabilities, answered by Big Data techniques
vehicle is loaded and unloaded, a dynamic such as complex event processing and geo
calculation of the optimal delivery sequence correlation. A real-time data stream is traced in
based on sensor-based detection of shipment order to assign shipments to available carriers,
items frees the staff from manual sequencing. based on their respective location and
On the road, telematics databases are tapped to destination. Interfaced through a mobile
automatically change delivery routes according application, crowd affiliates publish their current
to current traffic conditions. And routing position and accept pre-selected delivery
intelligence considers the availability and assignments.
The above two use cases illustrate approaches to Data analytics allow a comprehensive
optimizing last-mile delivery, yet they are assessment of customer satisfaction by merging
diametrically opposed. In both cases, massive multiple extensive data sources. For logistics
real-time information (originating from sensors, providers, this materializes in a combined
external databases, and mobile devices) is evaluation of records from customer touch
combined to operate delivery resources at points, operational data on logistics service
maximum levels of efficiency. And both of these quality, and external data. How do these pieces
Big Data applications are enabled by the fit together? Imagine the scenario of a logistics
pervasiveness of mobile technologies. provider noticing a customer who lowers
shipment volumes despite concurrently
publishing steady sales records through
3.3 Use Cases – Customer newswire. The provider then checks delivery
Experience records, and realizes that this customer recently
experienced delayed shipments. Looking at the
The aspect of Big Data analytics that currently bigger picture, this information suggests an
attracts the most attention is acquisition of urgent need for customer retention activity.
customer insight. For every business, it is vitally To achieve this insight not just with one
important to learn about customer demand and customer but across the entire customer base,
satisfaction. But, as organizations experience the logistics provider must tap multiple data
increased business success, the individual sources and use Big Data analytics. Customer
customer can blur into a large and anonymous touch points include responses to sales and
customer base. Big Data analytics help to win marketing activities, customer service inquiries,
back individual customer insight and to create and complaint management details. This digital
targeted customer value. customer trail is correlated with data from the
distribution network comprising statistical series
3.3.1 Customer value management on shipping volume and received service quality
levels. In addition, the Internet provides useful
Clearly, data from the distribution network customer insight: Publicly available information
carries significant value for the analysis and from news agencies, annual reports, stock
management of customer relations. With the trackers, or even sentiments from social media
application of Big Data techniques, and enriched sites enrich the logistics provider’s internal
by public Internet mining, this data can be used perspective of each customer.
to minimize customer attrition and understand From this comprehensive information pool, the
customer demand. logistics provider can extract the attrition
potential of every single customer by applying
Customer loyalty management techniques such as semantic text analytics,
natural language processing, and pattern
For most business models, the cost of winning a recognition. On automatically generated
new customer is far higher than the cost of triggers, the provider then initiates proactive
retaining an existing customer. But it is counter-measures and customer loyalty
increasingly difficult to trace and analyze programs. Although business relationships in
individual customer satisfaction because there logistics usually relate to the sender side, loyalty
are more and more indirect customer touch management must also target the recipient side.
points (e.g., portals, apps, and indirect sales Recipients are even more affected by poor
channels). Because of this, many businesses are service quality, and their feedback influences
failing to establish effective customer retention sender selection for future shipments. A good
programs. Smart use of data enables the example of this is Internet or catalog shopping:
identification of valuable customers who are on Recurring customer complaints result in the
the point of leaving to join the competition. Big vendor considering a switch of logistics provider.
But to include recipients into loyalty 3.4 Use Cases – New Business
management requires yet more data to be
processed, especially in B2C markets. Big Data
Models
analytics are essential, helping to produce an 3.4.1 B2B demand and supply chain
integrated view of customer interactions and
operational performance, and ensure sender
forecast
and recipient satisfaction. The logistics sector has long been a
5
macroeconomic indicator, and the global
Continuous service improvement
transportation of goods often acts as a
and product innovation
benchmark for future economic development.
The type of goods and shipped volumes indicate
Logistics providers collect customer feedback as
regional demand and supply levels. The
this provides valuable insight into service quality
predictive value of logistics data for the global
and customer expectations and demands. This
economy is constituted by existing financial
feedback is a major source of information for
indices measuring the macroeconomic impact of
continuous improvement in service quality. It is
the logistics sector. Examples are the Baltic Dry
also important input for the ideation of new
Index28, a price index for raw materials shipped,
service innovations. To get solid results from
and the Dow Jones Transportation Average29,
customer feedback evaluation, it is necessary to
showing the economic stability of the 20 largest
aggregate information from as many touch
U.S. logistics providers. By applying the power of
points as possible. In the past, the single source
Big Data analytics, logistics providers have a
of data has been ingests from CRM systems and
unique opportunity to extract detailed
customer surveys. But today, Big Data solutions
microeconomic insights from the flow of goods
provide access to gargantuan volumes of useful
through their distribution networks. They can
data stored on public Internet sites. In social
exploit the huge digital asset that is piled up from
networks and on discussion forums, people
the millions of daily shipments by capturing
openly and anonymously share their service
demand and supply figures in various
experiences. But extracting by hand relevant
geographical and industry segments.
customer feedback from the natural-language
content created by billions of Internet users is
Market intelligence for small and
like looking for that proverbial needle in a
medium-sized enterprises
haystack.
Sophisticated Big Data techniques such The aggregation of shipment records
as text mining and semantic analytics allow the (comprising origin, destination, type of goods,
automated retrieval of customer sentiment from quantity, and value) is an extensive source of
huge text and audio repositories. In addition, this valuable market intelligence. As long as postal
unsolicited feedback on quality and demand can privacy is retained, logistics providers can refine
be broken down by region and time. This enables this data in order to substantiate existing
identification of correlation with one-time external market research. With regression
incidents and tracking the effect of any initiated analysis, the fine-grained information in a
action. In summary, meticulous review of the shipment database can significantly enhance
entire public Internet brings unbiased customer the precision of conventional demand and
feedback to the logistics provider. This supply forecasts.
empowers product and operational managers to
design services capable of meeting customer
demand.
B2B demand and supply chain leverage data for better business performance.
forecast This is a good start, as solid use cases are a
8 fundamental requirement for adopting new
The result has high predictive value – and this information-driven business models. But there
compound market intelligence is therefore a needs to be more than a positive assessment of
compelling service that can be offered by third business value. The following five success
parties. To serve a broad range of potential factors must also be in place.
customers, the generated forecasts are
segmented by industry, region, and product
category. The primary target groups for 3.5.1 Business and IT alignment
advanced data services such as these are small
and medium-sized enterprises that lack capacity In the past, advancements in information
to conduct their own customized market management clearly targeted either a business
research. problem or a technology problem. While trends
such as CRM strongly affected the way sales and
Financial demand and supply chain service people work, other trends such as cloud
analytics computing have caused headaches for IT teams
attempting to operate dynamic IT resources
Financial analysts depend on data to generate across the Internet. Consequently, business units
their growth perspectives and stock ratings. and the IT department may have different
Sometimes analysts even perform manual perspectives on which changes are worth
checks on supply chains as the only available adopting and managing. But for an organization
source to forecast sales figures or market to transform itself into an information-driven
volumes. So for ratings agencies and advisory company – one that uses Big Data analytics for
firms in the banking and insurance sector, access competitive advantage – both the business units
to the detailed information collected from a and the IT department must accept and support
global distribution network is particularly substantial change. It is therefore essential to
valuable. An option for logistics providers is to demonstrate and align both a business case and
create a commercial analytics platform allowing an IT case for using Big Data (including
a broad range of users to “slice and dice” raw objectives, benefits, and risks). To complete a Big
data according to their field of research Data implementation, there must be a mutual
effectively creating new revenue streams from understanding of the challenges as well as a joint
the huge amount of information that controls commitment of knowledge and talent.
logistics operations. In the above use cases,
analytics techniques are applied to vast amounts 3.5.2 Data transparency and
of shipment data. This illustrates how logistics governance
providers can implement new information
driven business models. In addition, the Big Data use cases often build upon a smart
monetization of data that already exists adds the combination of individual data sources which
potential of highly profitable revenue to the jointly provide new perspectives and insights.
logistics provider’s top line. However, in many companies, the reality is that
three major challenges must be addressed to
ensure successful implementation. First, to
3.5 Success Factors for locate data that is already available in the
Implementing Big Data Analytics company, there must be full transparency of
information assets and ownership. Secondly, to
Our discussion of Big Data analytics has been prevent ambiguous data mapping, data
focused on the value of information assets and attributes must be clearly structured and
the way in which logistics providers can explicitly defined across multiple databases. And
thirdly, strong governance on data quality must 4.0 Way Forward
be maintained. The validity of mass query results
is likely to be compromised unless there are Looking ahead, there are admittedly numerous
effective cleansing procedures to remove obstacles to overcome (data quality, privacy, and
incomplete, obsolete, or duplicate data records. technical feasibility, to name just a few) before
And it is of utmost importance to assure high Big Data has pervasive influence in the logistics
overall data quality of individual data sources industry. But in the long run, these obstacles are
because – with the boosted volume, variety, and of secondary importance because, first and
velocity of Big Data – it is more difficult to foremost, Big Data is driven by entrepreneurial
implement efficient validation and adjustment spirit. Several organizations have led the way for
procedures. us – Google, Amazon, Facebook, and eBay, for
example, have already succeeded in turning
extensive information into business. Now we are
3.5.3 Data privacy beginning to see first movers in the logistics
sector. These are the entrepreneurial logistics
Data science skills providers that refuse to be left behind; the
opportunity-oriented organizations prepared to
A key to successful Big Data implementation is
exploit data assets in pursuit of the applications
mastery of the many data analysis and
described in this trend report. But apart from the
manipulation techniques that turn vast raw data
leading logistics providers that implement
into valuable information. The skillful application
specific Big Data opportunities, how will the
of computational mathematics makes or breaks
entire logistics sector transform into a data-
reliable and meaningful insights. In most
driven industry?
industries, the required mathematical and
What evolution can we anticipate in a
statistical skillset is scarce. In fact, a talent war is
world where virtually every single shipped item
underway, as more and more companies
is connected to the Internet? We may not know
recognize they must source missing data science
all of the answers right now. But this trend report
skills externally. Very specialized knowledge is
has shown there is plenty of headroom for
required to deploy the right techniques for each
valuable Big Data innovation. Joining resources,
particular data processing problem, so
labor, and capital, it is clear that information has
organizations must invest in new HR approaches
become the fourth production factor and
in support of Big Data initiatives.
essential to competitive differentiation. It is time
to tap the potential of Big Data to improve
Appropriate technology usage operational efficiency and customer experience,
and create useful new business models. It is time
Many data processing problems currently hyped
for a shift of mindset, a clear strategy and
as “Big Data challenges” could, in fact, have been
application of the right drilling techniques.
technically solved five years ago. But back then,
the required technology investment would have
shattered every business case. Now at a fraction
of the cost, raw computing power has
exponentially increased, and advanced data
processing concepts are available, enabling a
new dimension of performance. The most
prominent approaches are in-memory data
storage and distributed computing frameworks.
However, these new concepts require adoption
of entirely new technologies.
Reference
http:// https://www.journals.elsevier.com/big-
data-research
https://www.uts.edu.au/research-and-
teaching/our-research/global-big-data-
technologies-centre
https://www.businesswire.com/news/home/20
180430005735/en/Global-Big-Data-Market-
Research-Report-2018
https://journalofbigdata.springeropen.com/
http://www.bpic.asn.au/lci/about-
lca/potentialusesoflca
https://www.india.gov.in/official-website-
ministry-road-transport-and-highways
https://www.mckinsey.com/business-
functions/operations/our-insights/big-data-and-
the-supply-chain-the-big-supply-chain-analytics-
landscape-part-1
https://www.supplychain247.com/article/suppl
y_chain_big_data_analytics_innovation