Event Driven Supply Chain

Abstract:

Until just a few years ago, decision–making relied on paper-based processes, the human mind and a basic medium of communication. Today our decisions are based on what our systems reveal to us. But with greater information comes greater responsibilities. The larger the volume of information captured the longer the list of parameters for our managers to correlate and manage, which also means a greater risk of making mistakes. What if we had a mechanism to correlate mind-boggling volumes of information, reveal business insights directly to managers and make appropriate decisions automatically too?  Imagine the possible efficiencies in decision making that could be achieved.

This paper seeks to explore the application of Complex Event Processing technology across the Supply Chain through real –time correlation of events, hence reducing latency in decision making and maximizing business benefit.

Background:

The domain of Logistics has seen a significant transformational change over the last few years right from the way in which business is transacted, information is managed, or the way customers and partners respond to events happening across the Supply Chain. Technology has been the single strongest enabler with offerings such as enterprise applications, sensor edge technology, integration adapters, visibility solutions and concepts like web 2.0. Since the industry has grown primarily through consolidation, systems have been on-boarded inconsistently and the enterprise landscape has become a hotchpotch of multiple vendor applications and legacy systems. As a result, visibility across processes and lack of a single source of truth have emerged as major problems because systems are not well integrated and in some cases, not at all. The siloed nature of organizations has also contributed to this problem of lack of centralized control. Most Logistics Service Providers have different departments functioning as individual units, leading to a lack of process and IT architecture standardization.

Business Need:

The Logistics industry operates on wafer-thin margins, leaning heavily on Visibility, Predictability and Optimization in order to stay viable. Precision and accuracy of information in providing a single source of truth is of utmost importance in making impactful decisions.  The efficiency of the enterprise depends on how soon action is taken on events that occur across the enterprise

Some of the key drivers for an event driven supply chain include:

v�Situational Visibility across the operational and transactional landscape

There is a need for a real-time snapshot of operations and transactions across the enterprise. Often, dashboards provide an enterprise snapshot, but the retrospect nature of this picture diminishes the value. Information has a lifetime, and its value declines at a highly steep rate against time.�

v�Centralized control of supply chain operations

With service delivered across large regions and with operations in multiple locations, it becomes a challenge to track statistical parameters from an organization point of view. Often, strategic decision making is hindered due to lack of accurate numbers and parameters on operations across the diverse enterprise.

v�Putting together pieces of data and information in near-real-time to produce relevant business insights to managers

Making sense and correlating the massive volumes of business data and information that is captured across the enterprise is by any means an impossible task for logistics managers. Often many indicators revealing business threats are overseen due to the sheer volumes of information

v�Proactive responses to business situations and Exceptional Alerts

Any exception deserves an instant response. Delays in responding to critical situations are the major cause for customer dissatisfaction and operational inefficiencies. A robust alerting mechanism bringing to notice the origin of an exception helps in timely response to mitigating the impact on downstream operations.

Complex Event Processing: The Concept

Let’s take a look at what Complex Event Processing has to offer us. As the term suggests, it is the processing of complex events.�In business parlance, an Event is an occurrence of significance that one would want to track within the enterprise. There is a start and a definite end to every event. As much as “it is raining” is an event, a shipment delivery confirmation receipt is also an event. Associated with an event are attributes that define the event. In the case of the ‘raining’ event, the intensity of rain, the period for which it rained, the place over which it rained and the amount of rain are attributes that define that event. So then, what is a complex event?  A complex event is a situation where two or more events occur in different slices of time, from different sources but are related to each other in some way that is not easily comprehensible. Let’s us look at an example of a complex event that we may comes across in our daily life. Consider this:

v�You are driving on a freeway and it starts to rain (Event 1).

v�You then notice a ‘Slippery When Wet’ road sign, look at the speedometer and note that your speed is 75 Mph (Event 2).

v� Just then you recollect the fact that the speed limit on this road is 60 Mph (Event 3).

v�You reduce speed to 60 Mph to comply with the speed limit ( Event 4)

v�You also realize that your car has bad tyres and you should have taken her in for servicing last month ( Event 5)

v�Suddenly you feel your car skidding while you turn to take an Exit (Event 6).

v�Immediately you slow down to  40 Mph and drive more carefully because it is already raining , your tyres are bad, you’ve just experienced a skid and you don’t know what else can happen because you’ve missed getting your car serviced . ( Action 1)

 So the complex event consists of 6 events which are co-related in an intricate manner and reveal an insight when viewed together but may not mean much individually. In a business scenario, the application of Complex Event Processing technology provides similar type of insights and visibility through correlation of business events across time, thus bringing about tremendous efficiency in decision making.  

Some applications of CEP to Logistics are given below:

v� Vendor Managed Inventory (VMI): - Monitor Inventory threshold levels and allowing vendors to efficiently manage the movement of inventory of their clientele across the supply chain.

v�Dynamic re-planning: Automatic routing of critical shipments en-route to meet the delivery target.

v�Carrier Performance Monitoring: Monitoring multiple parameters of service provider (carriers) performance in real-time and appropriately tendering shipment to the right service provider.

v�Airlines and Airports : Allocation of gates and other resources based on arrival and departure times.

v�Hospitality :  Monitoring  demand based on website traffic rates and other channels of enquiries.

The Solution:  Advanced Control Tower (ACT):

The Advanced Control Tower is a product agnostic framework based on Complex Event Processing technology. It crunches large volumes of seemingly independent captured events occurring across multiple systems and derives and predicts a pattern based on insight, thereby automating decision making. Value is delivered through reducing latency at each point of the decision making cycle.

The ACT Framework is based on the following principles:

v�The ability of capturing information from different data-sources across the operational landscape.

v�Correlating and putting together the jig-saw puzzle of information to reveal an accurate and precise business picture.

v�Responding to this business picture through making the correct decision

v�Executing the decision with speed and perfection

Value is maximized only when the above four aspects are synthesized with minimal time latency. As information is captured, it has to be understood and utilized. The delay in capturing, understanding and responding to an activity or an incident is the challenge and the price that businesses pay in our times.

Components of ACT:

v�Scenario: A process all inclusive of events, business rules and resulting actions and alerts

v�Event: An occurrence of significance (operational or transactional)

v�Attributes: Elements of the event

v�Business Rules: The Logic that ties together the events and includes policies, requirements, and conditional statements that are used to determine the tactical actions that take place in applications and systems

v�Actions: The resulting processes or transactions that need to be triggered within applications across the enterprise or evoked from personnel across the organization

v�Alerts: Real-time ‘Notifications’ to management and operational personnel regarding events and actions

 Data is fed into the system in the form of ‘Events’ from across the landscape - IT systems, transactions and physical movements on the ground. These ‘Events’ are correlated using the business logic, and upon the occurrence of a pattern, an insight is revealed, provoking an action from a user (manually) or a system (automatically).�A set of seemingly disparate events when occurring together may signify something entirely different than from their individual occurrence.

There are two types of users who can utilize the framework – Operational Users and Business Users.

v�Operational users

·       �Configure Events and the associated attributes to be captured

·       �Configure business rules that vary with business need

·       �Select  Events to monitor, Business Rules to be applied, and Actions to be executed

·       �Configure alerts suitable for different levels of people and route them appropriately using the chosen medium

·       �Configure and view real-time dashboard elements providing insight into real-time operational parameters

v�Business Users

·       �Configure reports providing situational visibility to the state of operations

·       �Automate the generation of reports periodically or upon the occurrence of a pattern of events

·       �Configure alerts suitable for different levels of people and route them appropriately using the chosen medium

Benefits:

• Enhanced situational visibility into supply chain operations both at a strategic level and at a tactical operational level
• Faster and more accurate prediction of operational irregularities and exceptional occurrences
• Better utilization of information while it is still alive to deliver high impact
• Increased process automation for better customer satisfaction and optimized business activities