METHOD AND SYSTEM FOR DYNAMICALLY PREDICTING
ESTIMATED TIME FOR COMPLETION OF A CUSTOMER ORDER
FIELD OF THE INVENTION
[0001] The present invention relates to predicting time for completion of an
order received at an establishment. More particularly, the present invention relates
to a method and system to dynamically predict time of completion of an order
given by a user.
BACKGROUND OF THE INVENTION
[0002) An establishment or enterprise catering to provide services to
customers receives orders for various services from customers. The orders from a
customer may be received directly at a customer counter of the establishment or
via online systems using communication networks. Customer orders via online
systems have been increasing at a fast pace due to developments in
communication infrastructures and a common global market made accessible to
various establishments due to the online systems. Customer expectations for fast
and transparent service delivery have put increased pressure on service
establishments across the globe to upgrade their communication and service
delivery infrastructures. With particular reference to telecommunication and
mobile service providers, the ever increasing customer base, customer
expectations, and enhanced competition among the service provider have paved
way for more transparent, time bound and quality service delivery.
[0003] Once a customer places an order with a service provider, the order is
processed by the service provider and the customer is provided with the
service/product as per the received order. The customer is usually unaware of
inner working of the service provider systems and processes. However, the
customer is often anxious to know status of its order for the service. Enterprises
make several provisions for providing the order status to the customer. For
2
example, some enterprises keep the customer informed about time remaining for
the customer order to fmish or indicate by various means that the customer order
is ready. In a calVcontact centre environment, the customer is placed in a waiting
queue. The waiting queue is often managed by an automated system which keeps
the customer informed of the time remaining for its order to be taken up for
processing.
[0004] The automated systems are usually preconfigured and provide time
information to the customer based on some internal set values and algorithms as
per the queue position of the customer. However, as can be appreciated by a
person skilled in the art, service request load on the enterprise system is often
unpredictable. The timing information provided by the state of the art systems is
often way of the mark. Therefore, there exists a need for providing more reliable
and accurate timing information relating to orders placed at an enterprise to
customers.
OBJECTS OF THE INVENTION
[0005) It is an object of the present invention to provide a system to
dynamically predict time for completion of an order.
[0006] Another object of the present invention to provide a method for
dynamically predicting time for completion of an order.
[0007] Another object of the present invention to provide a system and
method for dynamically predicting time for order completion in large enterprises
with order completion invo Iving at least one node.
[0008] Another object of the present invention to provide a system and
method for dynamically predicting time for completion of complex order having
many subtasks in large enterprises.
3
[0009] Yet another object of the present invention to provide a system and
method for dynamically predicting time for completion of complex order having
varied traffic, choking, unavailability of nodes at different points in time.
SUMMARY OF THE INVENTION
[0010] This summary is provided to introduce concepts related to system and
methods for prediction of order completion time and the concepts are further
described below in the detailed description. However, it is intended neither to
identify all essential features of the present claimed invention nor for use in
determining or limiting the scope ofthe present claimed invention.
[0011] The present invention, in an embodiment, provides a method for
predicting a time of completion of an order placed at an estab lishment. The
method includes receiv ing an order from a user. The method dynamically predicts
the time of completion of the received order when the success rate of at least
predetermined number of previous order is greater than or equal to a
predetermined threshold during a predetermined time window. Further, the
method predicts the time for completion of the received order based on a static
value when the success rate of the at least predetermined number of previous
order is less than a predetermined threshold during the predetermined time
window. The method includes providing the predicted time for completion of the
received order to the user.
[0012] The present invention, in an embodiment, provides a computer
implemented system for predicting a time of completion of an order placed at an
establishment. The system includes a computer readable memory configured to
store computer executable instructions. Further, the system includes a data store
communicatively coupled to the computer readable memory and a processo r. The
data store is configured to store order processing data associated with the
establishment. The system also includes a processor communicatively coupled to
the memory and the data sto re. The processor receives computer executable
instructions from the memory. The processor is configured to receive the order
4
.,
from a user. The processor dynamically predicts the time of completion of the
received order when a success rate of at least predetermined number of previous
order is greater than or equal to a predetermined threshold during a predetermined
time window. The processor predicts the time for completion ofthe received order
based on a static value when the success rate of the at least predetermined number
of previous order is less than a predetermined threshold during the predetermined
time window. The processor provides the predicted time for completion of the
received order to the user.
[0013] In an embodiment, dynamically predicting the time for completion of
the received order includes retrieving historical time data and execution paths of
the at least one previous order from the data store.
BRIEF DESCRIPTION OF THE ACCOMPANYING DRAWINGS
[0014] The accompanying drawings, which are included to provide further
understanding of the invention and are incorporated in and constitute a part of this
specification, illustrate embodiments of the invention and together with the
description serve to explain the principles ofthe invention. In the drawings:
[0015] FIG. 1 illustrates an enterprise environment of a timing system for
dynamically predicting time for completion of an order placed at the enterprise,
according to an embodiment of the present invention;
[0016] FIG. 2 illustrates an order processing system communicatively
coupled to the timing system, according to an embodiment of the present
invention;
[0017] FIG. 3 illustrates an enterprise system communicatively coupled to the
timing system, according to an embodiment ofthe present invention;
[0018] FIG. 4 illustrates the timing system for dynamically predicting time
for completion of an order, according to an embodiment of the present invention;
and
5
[0019] FIGs. 5A-5C illustrate a method for dynamically predicting time for
completion of an order, according to an embodiment of the present invention.
DETAILED DESCRIPTION OF THE INVENTION
[0020] Exemplary embodiments of systems and methods for dynamically
predicting a time of completion of an order will now be discussed in further detail
with reference to FIGs. I -SC. It is to be understood that the systems and methods
described herein in accordance with the present invention may be implemented in
various forms of hardware, software, firmware, special purpose processors, or a
combination thereof. In particular, exemplary systems and methods described
herein may be implemented in software comprising program instructions that are
tangibly embodied on one or more program storage devices (e.g., hard disk,
magnetic floppy disk, RAM, CD ROM, DVD, ROM and flash memory), and
executable by any computing device or machine comprising suitable architecture.
It is to be further understood that because the constituent system modules and
method steps depicted in the accompanying figures can be implemented in
software, the actual connections between the system components (or the flow of
the process steps) may differ depending upon the manner in which the application
is programmed. Given the teachings herein, one of ordinary skilled in the related
art will be ab le to contemplate these and s imilar implementations or
configurations ofthe present invention.
[0021] Referring now to FIG. 1, an enterprise environment 100 of a timing
system 102 for dynamically predicting time for completion of an order placed at
an enterprise is illustrated, according to an embodiment of the present invention.
[0022] The enterprise environment 100 includes an order processing system
108, a timing system 102, and a user device 106, a ll of which are coupled to one
another via a communication network 104, accord ing to an embodiment of the
present invention.
6
[0023] An order from the user device I 06 is received by the order processing
system 108 via the communication network 104. The order processing system I 08
processes the order using one or more enterprise systems.
[0024] The order processing system 108 receives the order via the
communication network 104, determines the type and the execution path of the
received order. The type of the received order and the start time of processing of
the received order are also communicated to the timing system via the
communication network 104. The execution path is the collection of various
nodes that are required for the processing of the received order. The order is then
processed using one or more enterprise systems.
[0025] The order processing system I 08 may include a combination of one or
more systems that may, without intending any limitation, receive, acknowledge,
route, transform, manage, throttle, process, and monitor the order.
[0026] In an embodiment of the present invention, the order processing
system 108 may include at least one processing unit, storage, one or more input
devices, one or more output devices, one or more communication connections and
an interconnection mechanism which interconnects the component of the order
processing system 108. The processing unit executes computer readable
instructions and may be a real or a virtual processor.
[0027] The timing system I 02 retrieves data associated with an order being
processed by the order processing system I 08.
[0028] The timing system 102 predicts the time of completion ofthe received
order which is communicated to the user device I 06 via the communication
network 104. The timing system maintains and stores various data associated with
an order.
7
[0029] The timing system I 02 may include, but not limited to, at least one
timer to measure the time during the processing of order, at least one processing
unit, storage and an interconnection mechanism.
[0030] The user device 106 places a request for an order consisting of various
subtasks which is communicated via the communication network 104 for
processing in the order processing system I 08.
[0031] The communication network 104 facilitates the communication
between the components of enterprise environment 100. It conveys information
such as computer-readable instructions, compressed information, or other data.
[0032] In an embodiment of the present invention, the communication
network I 04 may inc lud e, but not limited to, a wired or wireless techniques
implemented with an electrical, optical, RF, infrared, acoustic, or other carrier.
[0033] In preferred embodiment the communication network 104 may
include a combination of one or more communication networks.
[0034] In an embodiment of the present invention, the user device 106 may
include a computing device, for example, but not limited to a ce llular mobile
phone, a smart phone, a pager, a Portable Digital Assistant (PDA), a laptop, and a
palmtop.
[0035] Referring to FIG. 2, the order process ing system 108 is illustrated,
according to an embodiment of the present invention. The order processing
system 108 includes an enterprise m iddleware 202, and an order entry system 204,
an order manager 206, and an enterprise system 208. The enterprise middleware
202 is communicatively coupled to the order entry system 204, the order_ manager
206, and the enterprise system(s) 208.
[0036] The enterprise middleware 202 is used for communication between
the elements of the order processing system 108. Enterprise m iddleware 202 is
a lways aware of order beginning, completion as well as status of a ll the tasks
8
going on in the order processing system 108. It also communicates the timing
system I 02 whenever an order or its state is passed between the systems.
[0037] The enterprise middleware 202 may include, but not limited to,
systems to manage order queues, to control the flow of orders, and to
communicate with various elements of the enterprise environment 100.
[0038] The order entry system 204 receives an order from the user device 106
via a communication network 104. It creates the order and submits it to the order
manager 206 for processing and fulfilment of the received order via the enterprise
middleware 202. Order entry system 204 reduces delay in order completion,
reduces error and allows order entry at any point of time.
[0039] According to an embodiment of the present invention, the order entry
system 204 may include, but not limited to, an interface which is configured to
receive the order and one or more decoders to create the order into a format which
is compatible with the order manager 206.
[0040] The order manager 206 understands the processing logic for an order
basis its type and orchestrates the fulfilment of order among dependent enterprise
systems 208. The order manager 206 divides the order into various subtasks and
determines the type of an order on the basis of nodes of enterprise system 208
required for the processing of the subtasks.
[0041] The enterprise system 208 is required to process the order in
completion or a part of it in the form of tasks defined in the order. The enterprise
system 208 has various nodes interconnected to each other, each responsible for
performing at least one subtask ofthe order.
[0042] According to an embodiment of the present invention, the enterprise
system 208 may include, but not limited to, at least one processing unit. The
processing unit executes computer-readable instructions and may be a real or a
9
virtual processor. In a multi-processing system, multiple processing units execute
computer-readable instructions to increase processing power.
[0043] The enterprise system 208 according to an embodiment of the present
invention is further illustrated in detail with reference to FIG. 3 below.
[0044] Referring to FIG. 3, the enterprise system 208 is illustrated, according
to an embodiment of the present invention. In one embodiment of the present
invention, Figure 3 illustrates execution paths of various orders processed through
the enterprise system 208. For instance, an order A is executed through node (A I)
304 and node (A2) 306. In the same way, order B is executed through nodes (Bl)
308, (B2) 310, (B3) 312 and in a similar way all the orders are executed through
this enterprise system 208.
[0045] Now for a received order R I execution path fol.lows the route of node
(l) 302, node (A2) 306, node (B2) 310, node (B3) 312. The order execution
timing for all the orders as well as for each node was stored in timing system 102
via enterprise middleware 202. An order similarity can be evaluated by comparing
the execution path of the received order with execution paths of previously
executed orders.
[0046] In one preferred embodiment of the present invention received order
R 1 is similar to order A as having one common node (A2) 306 and received order
R 1 is also similar to order B as having two common nodes (B2) 310 and (B3) 312.
In this way depending upon the similarity between the execution paths, similar
historical orders can be compared to the received order and their order completion
times can be fetched and stored in the timing system I 02 through enterprise
middleware 202.
[0047] FIG. 4 illustrates the timing system 102 for dynamically predicting
time for completion of an order, according to an embodiment of the present
invention.
10
[0048] The system 102 is illustrated in accordance with an embodiment of the
present invention. In one embodiment, the system 102 may include at least one
processor 110, an input/output (1/0) interface 112, and a memory 114. At least
one processor 110 may be implemented as one or more microprocessors,
microcomputers, microcontrollers, digita l s ignal processors, central processing
units, state machines, logic circuitries, and/or any devices that manipulate signals
based on operational instructions. Among other capabilities, at least one processor
110 is configured to fetch and execute computer-readable instructions stored in
the memory 114.
(0049] In another embodiment of the present invention, Figure 4 illustrates
the timing system 102 that is communicatively coupled to user device 106 and
order process ing system I 08 through a communication network 104. The timing
system I 02 consists of various sub modules like a time keeper 404 adapted to
monitor and log time taken by an order to be processed. The enterprise
middleware 202 communicates the status of the order to the time keeper 404
throu gh a time keeper queue 402. A data store 406 adapted to store information
related to order process ing. In an embodiment, the data store 406 may include a
time pattern repository or a database. A time predictor 414 adapted to predict
estimated time for completion of the order based on historic time data 4 10
maintained by the data store 406 as well as based on a current processing time and
an order type. The time pred ictor 414 estim ates completion time on the basis of
three modules that are historical time module 4 16, execution path module 418,
and special condition module 422 which are further dependent on the parameters
like historical time data, execution path, process ing load conditio ns and any
special conditions like any new laun ch of a new order or festiva l. By combining
all these parameters order completion time is calculated dynamically and
conveyed to the user throu gh communication means. FIGs. 5A-5C illustrate a
method for dynamically predicting time for completion of an order, according to
an embodiment ofthe present invention.
II
[0050] At step 502, an order from a user is received at an establishment. In an
embodiment, the order entry system 204 receives the order from the user device
106. The order entry system 204 may receive the order via the communication
network 104 from the user device 106. In an embodiment, a user may be
presented with a user interface ofthe user device 106, for example, a graphical
· user interface for placing an order. The information associated with the order may
be communicated to the order entry system 204 using a one or more
communication protocols. In an embodiment, a service executive may input order
directly into a user interface of the order entry system 204 on behalf of a
customer. In an embodiment, the order entry system 204 may be accessible only
by registered users via authentication mechanisms such as a password and
usemame. A new user may be registered with the order processing system 108
when placing an order for the first time. The registration may involve gathering
profile information and preferences ofthe user.
[0051] At step 504, a time window is determined for which the data
associated with previous orders is to be considered. In an embodiment, the timing
system 102 determines the time window to be used for considering data associated
with the previous orders. In an embodiment, the time window may be of the order
of seconds, minutes, hours, days, weeks, months, years or any other time duration.
[0052] At step 506, data associated with previous orders is retrieved. In an
embodiment, the timing system 102 retrieves the associated data from the data
store 406. The data retrieved may include execution path data 408, historical time
data 410, and other data 412.
[0053] The execution path data 408 may include, but not limited to, the data
related to the execution paths ofthe previously executed orders.
[0054] The historical time data 410 may include, but not limited to, the
averages of completion time of previous orders during at least predetermined
number of predetermined time window and the success and failure rates of
previously executed orders.
12
[0055] The other data 412 may include, but not limited to, a configurab le
value based on system load and performance, predetermined threshold, forecast
value and data related to occurrence of any special condition.
[0056] At step 508, the timing system 102 compares the success rate of
previous orders with a predetermined threshold during the predetermined time
window. A previous order is said to be a success when the previous order is
processed successfully within a predetermined time period. Similarly, when the
previous order is not completed in the predetermined time period it is considered
as failure in accordance with an embodiment of the present invention. The success
rate is determined by the number of orders processed successfully during a
predetermined time window. For example, if a total of 500 orders have been
rece ived in an hour and only 300 of those orders have been completed in
stipulated time, then the success rate is 60 per cent. The predetermined threshold
may be set by a system administrator based on historical data and associated
system performance and service request loads. If the success rate is of previous
orders is greater than or equal to a predetermined threshold, the method 500
proceeds to step 512 of FIG. 5B. If the success rate is of previous orders is less
than a predetermined threshold, the method 500 proceeds to step 510.
[0057] At step 510, when the success rate is less than the predetermined
threshold during the predetermined time window, the time for completion of the
received order is predicted based on a preconfigured static value. The static value
may be a time period preconfigured by a system administrator and stored in data
store. The time period signifies the time period in which a particular service/order
will be completed for a user based on the order. The preconfigured value may be
based on type of order. For example, for one type of order the preconfigured value
may be set to 5 minutes and for another type of order the preconfigured value may
be set to a 4 hours and so on.
[0058] At step 530, the preconfigured static value may be provided to the
user device 106 via the communication network 104 as the predicted time for
13
completion ofthe order. In an embodiment, the user may be informed via a short
messaging service, a push message, an email, on phone of the time of completion
ofthe order.
[0059] At step 512, the execution path ofthe received order is determined by
the order manager 206. The execution path is one or more nodes or subsystems of
the enterprise system 208 traversed for processing of the received order. The order
manager 206 divides the order into suborders and determines the processing logic
and the type of the order. Based on the type of order and existing business logic,
the execution path of the received order may be determined by the order manager
206.
[0060] In step 514, the order manager 206 may determine the execution paths
of previous orders which have been executed within the predetermined time
window. In one embodiment of the present invention, the execution paths of the
previous orders may be determined based on the execution path data 408.
[0061] At step 516, the execution path of the received order is compared with
the determined execution paths of the previous orders. At this stage, the method
500 proceeds to step 518 ofFIG. 5C.
[0062] At step 518, it is determined whether there is at least predetermined
n urn ber of previous order similar to the received order. Two orders may be
considered similar if they share a predetermined proportion of nodes in their
execution paths.
[0063] If there is at least predetermined number of previous order similar to
the received order, then the method 500 proceeds to step 520, otherwise to step
522.
[0064] At step 520, when at least predetermined number of previous order
similar to the received order is found, then a normal processing time for the
received order is determined based on the retrieved data associated with the
14
previous orders found similar to the received order. The normal processing time
may be determined based on a weighted average with more weight given to the
recent orders completed for the historical processing time for at least
predetermined number of previous order. In an embodiment of the invention,
normal processing time is obtained by choosing the latest available value among a
last hour average, a last day peak/ off peak average, a weekly average, a monthly
average and a forecast value. The last hour average is an average of the results of
the last hour. A day is divided into peak and off peak hours based on request load.
Last day peak average is calculated by the formula: Sum of data for peak
hours/total peak hours. Similarly, Last day off peak average is calculated by the
formu la: Sum of data for off peak hours/ total off peak hours. These formulae may
be used by built in functions to determine the said averages. Based on the time of
the day (peak or off peak hours) of the received order, the system selects whether
to compute Last day peak average value or last day off peak average value. The
weekly average is an average of the results of the last week and the monthly
average is an average of the results of the last month. The forecast value is an
output of time series forecasting.
[0065] At step 522, when no previous order similar to the received order is
found , then a normal processing time for the received order is determine based on
processing time of nodes in the execution time of the received order. l11e normal
processing time will be the processing time of at least one node in the execution
path ofthe received order in step 522.
[00661 In step 524, the presence of any special condition in the execution
path of the received order is checked. A special condition is determined in the
execution path of the received order based on a processing time deviation from a
normal processing time for at least one system placed in the execution path of the
received order. The special conditions may be a new launch, a festival or an event,
or planned downtime or maintenance of systems. l11e special conditions play a
vital role in dynamically predicting the estimated time for completion of a
customer order, as shown in step 526, on occurrence of the special condition, a
15
configurable value is added to the normal processing time (as determined in steps
520 and 522) associated with the execution path of the received order. In an
embodiment, the configurable value may be set by a system administrator. In a
case, there is no special condition, the normal processing time determined in steps
520 and 522 is taken as the final time of completion of the received order, as
shown in step 528.
[0067] The fmal step 530 ends the process by providing the predicted time for
completion of the received order to the user by means of the Communication
Network 104.
[0068] From the foregoing description, it will be apparent to one ordinarily
skilled in the art that many changes and modification can be made thereto without
departing from the spirit or scope of the invention as set forth herein.
[0069] Thus, it is intended that the scope of the present invention herein
disclosed should not be limited by particular disclosed embodiments described
above but should be determined only by a fair reading of complete specifications
and claims to fol low.

We claim:
1. A method for predicting a time of completion of an order placed at an
establishment having a computing environment and a processing unit (11 0), the
method comprising:
receiving, by an order entry system (204), the order from a user device
(I 06) operated by a user;
obtaining, by the processing unit (llO),the time of completion of the
received order when a success rate of at least predetermined number of previous
order is greater than or equal to a predetermined threshold during a predetermined
time window;
obtaining, by the processing unit (II 0), the time for completion of the
received order based on a static value when the success rate of the at least
predetermined number of previous order is less than a predetermined threshold
during the predetermined time window; and
providing, by the processing unit (II 0), the predicted time for completion
of the received order to the user.
2. The method as claimed in c laim I, wherein said receiving the order
comprises creating and submitting the order by the order entry system (204) to an
order manager (206) through an enterprise m iddleware (202) of the establishment.
3. The method as claimed in claim I, wherein said obtaining the time of
completion of the received order comprises retrieving historical time data and
17
execution paths of the at least predetermined number of previous order from a
data store.
4. The method as claimed in claim I, wherein said obtaining the time of
completion of the received order comprises determining a level of similarity
between the received order and the at least predetermined number of previous
order.
5. The method as claimed in claim 4, wherein said detennining the level of
similarity comprises comparing execution paths of the received order with the
execution paths of the at least predetermined number of previous order.
6. The method as claimed in claim 1, wherein said obtaining the time of
completion of the received order comprises determining a special cond ition in the
execution path of the received order based on a process ing time deviation from a
normal processing time for at least predetermined number of system placed in the
execution path of the received order.
7. The method as claimed in claim 6, wherein said normal processing time is
determined based on an average historical processing time for the at least
predetermined number of previous order.
8. The method as claimed in claim I, wherein said obtaining the time of
completion of the received order comprises adding a configurable va lue to the
18
normal processing time associated with the execution path of th~ received order
on occurrence the special condition.
9. A computer implemented system (1 02) for predicting a time of completion
of an order placed at an establishment, the system comprising:
a computer readable memory (114) configured to store computer
executable instructions;
a data store (406) communicatively coupled to the computer readable
memory (114) and a processor (11 0), the data store (406) configured to store order
processing data associated with the establishment; and
the processor (110) communicatively coupled to the memory (114) and the
data store (406) receiving computer executable instructions from the memory
(114), the processor (110) configured to:
receive the order from a user device (I 06) operated by a user;
obtain the time of completion of the received order when a success
rate of at least predetermined number of previous order is greater than or
equal to a predetermined threshold during a predetermined time window;
obtain the time for completion of the received order based on a
static value when the success rate of the at least predetermined number of
previous order is less than a predetermined threshold during the
predetermined time window; and
provide the predicted time for completion of the received order to
the user.
19
10. The system as claimed in claim 9, wherein the processor (11 0) is further
configured to retrieve historical time data and execution paths of the at least
predetermined number of previous order from the data store (406).
II. A method for predicting a time of completion of an order placed at an
establishment substantially as herein described with reference to the
accompanying drawings.
12. A computer implemented system for predicting a time of completion of an
order placed at an establishment substantially as herein described with reference
to the accompanying drawings.