Cognitive Procurement

The intelligent order concierge system solves the problem of order delays and errors in the existing procurement process by monitoring and analyzing procurement data sources, predicting and repairing order problems, and improves procurement efficiency and reliability.

CN111768207BActive Publication Date: 2024-12-03ACCENTURE GLOBAL SOLUTIONS LTD
0 Cites 0 Cited by

Patent Information

Application Number
CN202010219728.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Priority Date
2019-03-27
Filing Date
2020-03-25
Publication Date
2024-12-03
Estimated Expiration
2040-03-25

AI Technical Summary

Technical Problem

The existing procurement process is susceptible to order delivery errors or exceptions, resulting in order delays or cancellations, and lack of mechanisms to effectively manage exceptions, resulting in inefficiency and commercial impact of the procurement process.

Method used

The intelligent order concierge system is adopted to monitor and analyze procurement-related data sources, build behavioral models, predict potential events and automatically repair order problems, and ensure the smooth fulfillment of orders.

Benefits of technology

Improve order tracking and management efficiency, reduce potential delays and costs, enhance the agility and reliability of the procurement process, and reduce customer costs and delay risks caused by the supplier.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN111768207B_ABST
    Figure CN111768207B_ABST
Patent Text Reader

Abstract

Embodiments of the present disclosure relate to cognitive procurement. An example of cognitive procurement is described. In an example embodiment, procurement-specific data sources associated with at least one of a process, organization, and industry related to procurement operations are monitored. From the monitored procurement-specific data, operational behavior patterns are identified. Subsequently, a behavioral model of an order is constructed using the operational behavior patterns and a pre-existing behavioral model library. A procurement interaction is received from a user indicating an inquiry for processing an order. The order is tracked through a cognitive order concierge. Using the behavioral model, potential events related to the order are predicted, the potential events indicating problems that affect the order. Therefore, the problems that affect the order are proactively repaired to automatically troubleshoot the order. In the example, the user is notified based on the repair requirements.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] Embodiments of the present disclosure relate to cognitive purchasing. Background Art

[0002] Recently, the importance of procurement as a key process has increased significantly. Initially, procurement began as a way to integrate purchasing into supply chain management when most large companies were trying to manage their operating costs. Currently, the procurement process begins with a purchaser making a purchase request or placing an order. In procurement operations, after the order has been placed, multiple steps are performed to verify the order and track the order. Once verified, the order can be sent to the penultimate step of order preparation, that is, for assembling the ordered product or for providing the ordered service, as the case may be. Subsequently, the order is delivered to the customer through shipping in the case where the order is for a product, or, in the case where the order is for a service, the order is delivered to the customer through activation.

[0003] The existing source-to-pay or purchase-to-pay process or sales operation for any organization usually starts with the preparation of a purchase application. The purchase application can cause multiple suppliers to submit quotations in response to RFQ (request for quotation) or RFP (request for solution). An assessment can be performed to review each RFQ or RFP for their eligibility and ability to meet the criteria specified in the purchase application. The assessment of each RFQ or RFP can lead to making a purchasing decision and selecting a supplier. A purchase order can be issued to the supplier. The supplier may now be required to handle the manufacturing (if necessary) and logistics of the goods according to the purchase application. After the logistics of the goods according to the purchase application has been completed, a cost list or invoice will be issued by the supplier. The organization will receive the goods according to the purchase application and will continue to verify that the received goods can be consistent with the purchase application. When the information is found to be accurate and all parameters are in good condition, the cost list will be sent to the payment department.

[0004] Thus, the sales operation process has multiple milestones, and each milestone is susceptible to errors or problems that may affect an order. For example, steps in a sales operation that may involve logistics and supply chain activities may be particularly susceptible to errors and exceptions that may affect the delivery of an order. These errors and exceptions may include, for example, the unavailability of a product or the occurrence of an event that may result in the delay or cancellation of an order or problems with the delivery of an order.

[0005] Thus, with current models for automated procurement, exceptions are typically detected when the process reaches a milestone affected by the exception, thereby leaving little room and time to manage the exception in an effective manner. As a result, orders are delayed or cannot be fulfilled, and such order processing and management can adversely affect the business. Thus, a technical problem with currently available procurement processes is that they may be ineffective, inaccurate, and / or non-scalable.

[0006] There is a need for systems and methods that can increase the spontaneity and agility of tracking orders for their potential delivery delays, especially those focused on value-critical ones. There is also a need to reduce the risk of customer costs and delays caused by recent supplier deliveries. In addition, there is a need for a procurement concierge that can enable optimization of a continuous procurement paradigm that includes continuous sensing of emerging issues, exceptions, emerging risks and opportunities, evaluation of recommendations, improved delivery assurance, increased agility in tracking orders for their potential delivery delays, and increased productivity in communication with suppliers in such order tracking to negotiate the best possible outcome. Summary of the invention

[0007] According to one or more example embodiments, a system is disclosed, comprising: a processor; a behavior model builder, which is coupled to the processor, the behavior model builder comprising: a behavior identifier, which is used to: monitor a procurement-specific data source associated with at least one of a process, an organization, and an industry related to procurement operations; and identify operational behavior patterns from the monitored procurement-specific data, and a modeler, which is used to: use the operational behavior patterns and use a pre-existing behavior model library to build a behavior model for an order; and associate the order with at least one of the procurement-specific data sources; a monitor, which is coupled to the processor, the monitor is used to: receive procurement interactions from a user, the procurement interactions are used to indicate a behavior sent by the user to the system for Processing queries for orders; and facilitating the placement of orders associated with at least one of a process, organization, and industry related to procurement operations; and an updater, coupled to the processor, the updater comprising: an order tracker, which is used to interact with a procurement-specific data source for retrieval of the status of an order; a problem predictor, comprising: a model deployer, which is used to deploy a behavioral model to predict potential events related to a received order to be fulfilled using procurement interactions based on the behavioral model, wherein the potential events indicate problems affecting the order; and a parser, which is used to proactively fix problems affecting the order to automatically troubleshoot the order; and a dialog manager, which is used to facilitate user interaction with the procurement-specific data source according to the requirements of the problem affecting the order.

[0008] According to one or more example embodiments, a method is disclosed, comprising: monitoring procurement-specific data sources associated with at least one of processes, organizations, and industries related to procurement operations; identifying operational behavior patterns from the monitored procurement-specific data; using the operational behavior patterns and using a pre-existing behavioral model library to build a behavioral model for an order; associating the order with at least one of the procurement-specific data sources; receiving procurement interactions from a user, the procurement interactions being used to indicate queries sent by the user to the system for processing an order; facilitating the placement of an order associated with at least one of the processes, organizations, and industries related to procurement operations; interacting with the procurement-specific data source for retrieval of the status of the order; predicting potential events related to the received order to be fulfilled using the procurement operation based on the behavioral model, wherein the potential events indicate problems that affect the order; proactively repairing the problems that affect the order to automatically troubleshoot the order; and facilitating the user's interaction with the procurement-specific data source as required by the problems that affect the order.

[0009] According to one or more example embodiments, a non-transitory computer-readable medium is disclosed, comprising machine-readable instructions executable by a processor to: monitor procurement-specific data sources associated with at least one of processes, organizations, and industries related to procurement operations; identify operational behavior patterns from the monitored procurement-specific data; use the operational behavior patterns and a pre-existing behavioral model library to build a behavioral model for an order; associate the order with at least one of the procurement-specific data sources; receive procurement interactions from a user, the procurement interactions indicating queries sent by the user to the system for processing an order; facilitate the placement of an order associated with at least one of the processes, organizations, and industries related to procurement operations; interact with the procurement-specific data source for retrieval of the status of the order; predict potential events related to the received order to be fulfilled using the procurement operation based on the behavioral model, wherein the potential events indicate problems that affect the order; proactively repair problems that affect the order to automatically troubleshoot the order; and facilitate user interaction with the procurement-specific data source as required by the problem that affects the order. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Features of the present disclosure are illustrated by examples shown in the following drawings. In the following drawings, similar reference numerals indicate similar elements, in the drawings:

[0011] Figure 1 illustrates a schematic block diagram of a smart order concierge system according to an example embodiment of the present disclosure;

[0012] Figure 2illustrates components of a smart order concierge system for the purpose of providing automated order tracking and issue resolution according to an example embodiment of the present disclosure;

[0013] Figure 3 illustrates an overall view of an order behavior model and an environment behavior model according to an example embodiment of the present disclosure;

[0014] Figure 4 The present invention illustrates an exemplary embodiment of the present invention. Figure 3 Order behavior model;

[0015] Figure 5 The present invention illustrates an exemplary embodiment of the present invention. Figure 3 Environmental behavior model;

[0016] Figure 6 illustrates monitoring and automated investigation of orders according to an example embodiment of the present disclosure;

[0017] Figure 7 illustrates an example of conversation management for remediation of issues related to fulfillment of an order according to an example embodiment of the present disclosure;

[0018] Figure 8 illustrates hypothesis generation and evidence collection for remediation of problems related to fulfillment of orders according to other example embodiments of the present disclosure;

[0019] Fig. 9 The present invention is based on an exemplary embodiment of the present disclosure. Figure 4 Example of next order prediction using the order behavior model;

[0020] Fig.10 illustrates an example of a fix for a problem related to fulfillment of an order according to an example embodiment of the present disclosure;

[0021] Fig.11 illustrates an example of ongoing inventory maintenance resulting from deploying a smart order concierge according to an example embodiment of the present disclosure;

[0022] Fig.12 illustrates a hardware platform for implementing a system according to an example embodiment of the present disclosure;

[0023] Fig.13 A method of automated order troubleshooting according to an example embodiment of the present disclosure is illustrated. DETAILED DESCRIPTION

[0024] For simplicity and illustration purposes, the present disclosure is described by mainly referring to its examples. The examples of the present disclosure described herein can be used together according to different combinations. In the following description, details are set forth in order to provide an understanding of the present disclosure. However, it will be apparent that the present disclosure can be put into practice without restriction to all these details. In addition, throughout the present disclosure, the terms "one" and "an" are intended to represent at least one of a particular element. As used herein, the term "includes" means including but not limited to, and the term "including" means including but not limited to. The term "based on" means at least partially based on, the term "based upon" means at least partially based on, and the term "such as" means such as but not limited to.

[0025] The present disclosure describes various aspects related to automated order troubleshooting and intelligent purchase order processing. In one example, the aspects described in the present disclosure facilitate processing an order after it has been placed by a customer and received at a service provider, or in the first place in a manner that addresses clearly perceived potential requirements and factors. For purposes of the present disclosure, the term customer is used to indicate any party that places an order, and the term service provider has been used to indicate any party that sells, manufactures, or otherwise provides goods, products, and services to a customer.

[0026] From the customer's perspective, the technology of the present disclosure provides customized orders that fully meet their requirements while taking into account industry trends and practices. At the same time, for example, any social, economic or political conditions that may affect business, industry trends and therefore orders are taken into account while placing orders. In addition, the present disclosure provides these from the perspective of the service provider. For example, given industry norms, the service provider may be able to effectively adapt to customer requirements, keeping in mind various factors (past, present and future) that may affect the fulfillment of an order or the need to generate an order.

[0027] First, as an example, from an implementation perspective, the technology for the Smart Order Concierge is embodied in two phases. First, the behavioral model captures operational behavior patterns that indicate procurement-specific behaviors associated with a process, as well as an organization and industry, or a combination thereof. In other words, the behavioral model captures operational behavior patterns followed by customers associated with a specific process, or by a customer organization, or by the industry to which the customer belongs. In the second phase, the behavioral model is deployed when an order is placed or received, as the case may be. The behavioral model is deployed to assess past, present, and future scenarios to determine whether an order may be affected, and if it is likely to be affected, automatically troubleshoot the order to mitigate or completely prevent the order from being affected.

[0028] In operation, as part of model building, procurement-specific data sources associated with processes, organizations, industries, or combinations thereof related to procurement operations may be monitored. In an example, processes, organizations, and industries related to procurement operations may be selected based on the operations of a party implementing the techniques of the present disclosure. For purposes of the present disclosure, the term "procurement operations" is intended to encompass sales operations or procurement operations performed by service providers in the form of selling products and services and operations performed by customers in the form of placing orders for products and services provided by various service providers, sellers, and manufacturers.

[0029] For example, if the party implementing the smart order concierge technology of the present disclosure is a service provider, the process or organization or industry can be related to the customers catered by the service provider or the industry or field in which the service provider operates. On the other hand, for example, if the party implementing the smart order concierge is a customer, the monitoring of the process, organization, and industry can be related to the industry or field in which the customer operates or the standard industry-specific process that the customer will implement. Monitoring of sales operations can be achieved in real time by querying various databases and repositories of information online and offline (from which relevant data can be retrieved). For example, social media, online portals, and other websites that carry information about processes, organizations, industries, or combinations thereof can be crawled for retrieving and monitoring the above-mentioned information.

[0030] Once the procurement-specific data has been retrieved and stored, operational behavior patterns may be identified based on the procurement-specific data. In an example, the behavioral patterns so identified may indicate current and future trends and differences between current and previous trends that may affect, for example, demand and supply in an industry based on the procurement-specific data. The operational behavior patterns may be identified using the procurement-specific data based on predefined rules that reflect processes, organizations, and industries related to a party executing a smart order concierge. In an example, the predefined rules may be based on examples of historical records of the order process in the industry, process, organization, and the like.

[0031] With the operational behavior pattern determined, a behavioral model of the order is then constructed to simulate the monitored procurement-specific data and the operational behavior pattern as mentioned above. In an example, a pre-existing behavioral model library can be the basis for constructing the behavioral model as well as the operational behavior pattern. For example, a pre-existing behavioral model library can be used as a template for generating a behavioral model that attempts to mimic the operational behavior pattern. According to one aspect, an order is associated with at least one of the procurement-specific data sources, which in turn associates the behavioral model with the procurement-specific data source. In one aspect, the smart order concierge receives a procurement interaction from a user. The procurement interaction is used to indicate a query sent by a user to the system for processing an order. The smart order concierge can use the behavioral model and the procurement interaction to facilitate the placement of orders associated with at least one of the processes, organizations, and industries related to procurement operations. The first phase of the smart order concierge is concluded with the construction of the behavioral model and the placement of the order.

[0032] As mentioned previously, in the second phase, behavioral models are implemented for order management and, if necessary, for automatically repairing orders. Automation of repair implies that there is no human intervention at any stage, and the process of order processing, management, and troubleshooting is machine-performed. In the second phase, behavioral models and associated processes are triggered as soon as an order is placed to accurately ensure problem resolution. The Smart Order Concierge tracks procurement operations from order placement to final delivery for the ability to predict potential events. In an example, the Smart Order Concierge may include an order tracker to interact with procurement-specific data sources for retrieving the status of an order.

[0033] As part of the deployment of the behavioral model, any potential events related to the received orders to be fulfilled using procurement operations are predicted based on the behavioral model. Potential events indicate problems that can directly or indirectly affect the order. For example, a direct impact can be in the form of a real-world event (such as a natural disaster) that can affect the fulfillment of the order. Indirect impacts can be in the form of economic, political, or social conditions brewing in a region that can affect demand or supply and potentially affect orders. In addition, as previously mentioned, the prediction of affected orders can be from the perspective of the service provider or the consumer. In either case, the factors can be the same or similar. As previously mentioned, the behavioral model is constructed to track procurement operations from order placement to final delivery to predict potential events.

[0034] In the example, as part of the prediction of potential events, an investigation of exceptions previously encountered during the procurement process and problems previously generated during the procurement process may also be completed to identify potential events or problems that affect the order. In addition, the present disclosure provides for the aggregation of newly discovered influencing factors in a behavioral model. Therefore, in one example, an order is investigated to determine whether the order conforms to the behavioral model, which, as mentioned above, is done based on the operational behavior pattern associated with the order. In the event that the order does not conform to the behavioral model, a hypothesis for the order is generated and incorporated into the behavioral model to enable prediction of potential events in such circumstances. In other words, if the order does not conform to the behavioral model, which means that the behavioral model cannot accurately reflect the operational pattern, the new behavioral pattern is incorporated into the behavioral model.

[0035] If an event is predicted that could potentially impact an order, automated troubleshooting of the event can be initiated and proactive fixes to the problem that impacts the order can be implemented. For example, in one case, a hypothesis is generated based on historical data, and recommendations for resolving the problem are provided based on the hypothesis. Additionally, as part of the automation of order remediation, alerts to all users (such as stakeholders, such as the sales team, operations team, delivery team, and verification team) can be generated in sales operations in response to predictions of potential events that in turn indicate a problem that impacts an order. This can allow users to proactively provide solutions to identified problems.

[0036] Figure 1 A smart order concierge environment 110 according to an example embodiment of the present disclosure is illustrated, which may include a system 110 for providing automated order troubleshooting. For purposes of the present disclosure, the smart order concierge environment 110 is referred to as the system 110. In an example embodiment, the system 110 facilitates processing orders after the order has been placed by a customer and received at a service provider, or in a manner that addresses potential requirements and factors that are not obviously perceptible. Therefore, the system 110 can be deployed at the client or at the service provider end. As also mentioned previously, although the term procurement operation has been used before and will be used later, it should be understood that the term is intended to cover sales operations or procurement operations performed by service providers in the form of selling products and services and operations performed by customers in the form of placing orders for products and services provided by various service providers, sellers, and manufacturers.

[0037] At the client, in the example, system 110 can provide tailored orders that fully meet their requirements, taking into account industry trends and practices. At the same time, system 110 can also consider external factors while placing an order, which may be hidden factors, for example, any social, economic or political situation that may affect business, industry trends, market demand and therefore affect the order. In the example, once the order has been fulfilled, system 110 can facilitate payment. System 110 can facilitate the dialogue between the customer and the service provider according to the requirements of the order. System 110 generates automated communications with the customer or supplier after predicting any problems that affect the order. In addition, system 110 can provide similar equipment when being implemented at the terminal of the service provider. For example, system 110 can be able to consider industry specifications and various other factors that may affect the fulfillment of the order or affect the demand that first caused the order, so as to allow the service provider to effectively adapt to customer requirements.

[0038] As part of implementing the above functions, the system 110 may include a processor 120, a behavior model builder 130, a monitor 140, and an updater 150. The processor 120 may be coupled to the behavior model builder 130, the monitor 140, and the updater 150.

[0039] In an example, the behavior model builder 130 may build a behavior model that may reflect various factors that may affect an order, and the updater 150 may use the behavior model and provide automated order repair. In addition, the updater 150 may also include a configuration for improving the behavior model if the behavior model cannot adapt to the situation. The behavior model builder 130 and the updater 150 may include other components that are Figure 2 Explain in detail.

[0040] According to an embodiment of the present disclosure, the behavior model builder 130 can be configured to monitor a procurement-specific data source associated with at least one of a process, an organization, and an industry related to procurement operations. The behavior model builder 130 can also identify operational behavior patterns from the monitored procurement-specific data. The operational behavior pattern can be used and a pre-existing behavior model library can be used to build a behavior model for an order. In an example, a pre-existing behavior model library can be a basis for building a behavior model and an operational behavior pattern. For example, a pre-existing behavior model library can be used as a template for generating a behavior model that attempts to mimic an operational behavior pattern. An order can be associated with at least one of the procurement-specific data sources. In an example, the procurement-specific data source also includes at least one of the following: a supplier list, features, benefits, costs, availability, location, delivery methods for goods, services, and information, innovation assessments, risk assessments, technical assessments, supplier collaboration from receiving supplier data, and the like.

[0041] In an example, a procurement interaction may be received from a user, for example, by monitor 140. The procurement interaction may indicate a query sent by a user to system 110 for processing an order. In an example, an order may include a purchase of goods and services. Monitor 140 may be configured to facilitate the placement of an order. In an example, an order may be associated with at least one of a process, an organization, and an industry related to procurement operations.

[0042] The system 110 may be configured such that the updater 150 may be configured to interact with a procurement-specific data source to retrieve the status of an order. The updater 150 may track an order from the placement of the order to final delivery. The updater 150 may be configured to deploy a behavioral model to predict potential events related to a received order to be fulfilled using procurement interactions based on the behavioral model, wherein the potential events indicate problems affecting the order. In other words, the updater 150 may facilitate that the behavioral model may attempt to capture as many failure points as possible in the sales-to-delivery process in order management. In an example, the updater deploys a behavioral model built by the behavioral model builder 130 through the processor 120. The updater 150 may be configured to proactively fix problems affecting orders to automatically fix the order. In an example, the updater 150 facilitates real-time resolution of problems affecting orders on the zero (0) day state. According to an embodiment of the present disclosure, the updater 150 may be configured to facilitate user interaction with a procurement-specific data source according to the requirements of the problem affecting the order. In an example, the updater 150 may be configured to resolve problems affecting orders on the 0th day from the expense list generated for the order and obtain a payment commitment before the due date. In an example, the updater 150 can proactively guide users of the system 110 in anticipation of potential needs. Users of the system 110 can be service providers, employees of organizations with purchasing requirements, customers of various goods and services, suppliers, etc. The updater 150 can be configured to contact providers before inquiries and complaints are generated, so as to effectively resolve, repair or correct any problems affecting orders before such problems arise.

[0043] Figure 2Components of an intelligent order concierge system for providing automated order tracking and problem solving according to an example embodiment of the present disclosure are illustrated. According to an embodiment of the present disclosure, the behavior model builder 130, the monitor 140, and the updater 150 may also include various components to perform the techniques described by the system 110. The behavior model builder 130 may include a behavior identifier 210 and a modeler 220 to facilitate the behavior model builder 130 to complete the various functions of the system 110. In addition, the updater 150 may include an order tracker 230, a problem predictor 240, a dialogue manager 250, and a model enhancer 280. The problem predictor 240 may also include a model deployer 260 and a parser 270 to facilitate the updater 150 to complete the various functions of the system 110.

[0044] In an example, from an operational perspective, the system 110 is configured such that the behavior model builder 130, the monitor 140, and the updater 150 operate synchronously through the processor 120. The system 110 may be configured such that the techniques described herein are implemented in three phases.

[0045] In a first phase, the behavioral model builder 130 builds a behavioral model that captures operational behavior patterns indicative of procurement-specific behaviors associated with a process, organization, and industry, or a combination thereof, associated with a party implementing the system 110. In a second phase, the monitor 140 receives procurement interactions from a user, the procurement interactions being indicative of queries sent by the user to the system for processing an order. The monitor 140 can facilitate the placement of orders associated with at least one of the processes, organizations, and industries related to procurement operations. Then, in a third phase, as the case may be, the updater 150 deploys the behavioral model when an order is placed or received. In an example, the updater 150 deploys the behavioral model to predict the behavior associated with the received order to be used through Figure 1 In an example, the updater 150 deploys a behavioral model to assess whether a problem affecting an order may be related to the manufacturing of the order if the order requires the manufacturing of a product before delivery. In an example, the updater 150 deploys a behavioral model to assess whether a problem affecting an order may be related to the shipping of the order after the order has left the service provider site and is in transit to the customer site. Therefore, the updater 150 can initiate an automatic fix for the problem to mitigate or completely prevent the order from being affected.

[0046] In one aspect, the behavior model builder 130 may be coupled to the processor 120. The behavior model builder 130 may include a behavior identifier 210 and a modeler 220. The behavior identifier 210 may be configured to monitor a procurement-specific data source associated with at least one of a process, an organization, and an industry related to procurement operations. The behavior identifier 210 may be configured to identify operational behavior patterns from the monitored procurement-specific data.

[0047] In operation, as part of model building, the pattern identifier 102 can monitor various information sources in real time that can be relevant to maintaining attention to factors that may be able to affect the completion or non-completion of orders, whether for service providers or customers. Therefore, the behavior identifier 210 can monitor procurement-specific data sources associated with processes, organizations, industries, or combinations thereof related to procurement operations that can be monitored. In an example, the processes, organizations, and industries related to procurement operations can be selected based on the operations of a party implementing the technology of the present disclosure.

[0048] For example, if the implementer of system 110 is a service provider, the process or organization or industry can be related to the clients catered by the service provider or the industry or field in which the service provider operates. On the other hand, for example, if the implementer of system 110 is a client, the monitoring of processes, organizations, and industries can be related to the industry or field in which the client operates or standard industry-specific processes that the client will implement. The behavior identifier 210 can query various databases and repositories of information (from which relevant data can be retrieved) in real time online and offline. For example, the behavior identifier 210 can query social media, online portals, and other websites that carry information about processes, organizations, industries, or combinations thereof, which can be captured by the behavior identifier 210 for retrieving and monitoring the above-mentioned information.

[0049] Once the sales-specific data has been retrieved and stored, the behavior identifier 210 may identify operational behavior patterns based on the procurement-specific data. In an example, the behavior patterns so identified by the behavior identifier 210 may indicate fluctuations in the manufacturing paradigm of a particular service provider. In an example, the behavior patterns so identified by the behavior identifier 210 may indicate fluctuations related to shipping and transportation for a particular product, service provider, etc. The behavior identifier 210 may use the procurement-specific data to identify operational behavior patterns based on predefined rules, which may attempt to simulate processes, organizations, and industries related to the implementation of smart orders and purchase order concierges by implementers. In an example, the predefined rules may be based on cases for the order process in the industry, process, organization, and historical records for the implementer.

[0050] With respect to the determined operational behavior pattern, the modeler 220 may use the operational behavior pattern and use a pre-existing behavioral model library to construct a behavioral model for the order. In an example, the modeler 220 may consider a pre-existing behavioral model library as a basis for constructing a behavioral model. For example, the modeler 220 may use a pre-existing behavioral model library as a template for generating a behavioral model that attempts to mimic the operational behavior pattern. In an example, an order may be a service requested by a customer. In an example, an order may be a request for manufacturing a particular type of goods. In an example, an order may be a new order placed by a new customer to a new service provider. In an example, an order may be an ongoing contract that a customer may have with a service provider for the delivery of goods and services at a specified time. In an example, the modeler 220 may be configured to associate an order with at least one of the procurement-specific data sources.

[0051] In one aspect, the monitor 140 may be coupled to the processor 120. The monitor 140 may receive a procurement interaction from a user, the procurement interaction being used to indicate a query sent by the user to the system for processing an order. The procurement interaction may indicate a query sent by the user to the system for processing the purchase of goods and services. In an example, the user's procurement interaction also includes at least one of the following: screening historical procurement requests, investigating cost fluctuations, developing requests for solutions, tracking promotions, establishing procurement recommendations, monitoring compliance with procurement guidance, and monitoring the acquisition of goods, services, or information. The monitor 140 may be configured to facilitate the placement of orders associated with at least one of a process, organization, and industry related to procurement operations.

[0052] In an example according to the present disclosure, the behavior model constructed by the behavior model builder 130 may be an order behavior model. Figure 3 and Figure 4 The model is explained in detail. The order behavior model may include elements of order behavior after the order is placed, such as product expiration, product receipt, billing issues, payment settlement, etc. In an example according to the present disclosure, the behavior model constructed by the behavior model builder 130 may be an environmental behavior model. Figure 3 and Figure 5 Explain the model in detail. The environmental behavior model may include elements of environmental behavior after placing an order, such as shipping, manufacturing, transportation, payment, etc. In an example according to the present disclosure, the behavior model constructed by the behavior model builder 130 may be a combination of an environmental behavior model and an order behavior model.

[0053] It should be understood that the behavior model so constructed also takes into account industry knowledge to accurately represent the operational behavior patterns related to a specific industry. For example, the range of industry types can include the financial industry, the pharmaceutical industry, the oil and gas industry, and the retail industry. In each industry, the work protocol or workload can be specific to each industry, and therefore the demand for different behavior models is specific. For example, in the pharmaceutical industry, the resource-intensive workload can include gene sequencing workload, personal medicine workload, adverse drug reaction capture workload, and compliance workload. In another example, in the oil and gas industry, the resource-intensive workload can include oil and gas exploration workload and seismic migration workload. In yet another example, in the retail industry, the resource-intensive workload can include cross-selling or up-selling recommendation workload, supply chain optimization workload, post-payment audit workload, and customer joining workload.

[0054] In one aspect, the system 110 includes an updater 150 coupled to the processor 120. The updater 150 may also include an order tracker 230, a question predictor 240, a dialog manager 250, and a model enhancer 280. The question predictor 240 may also include a model deployer 260, a parser 270 to facilitate the updater 150 to complete various functions of the system 110.

[0055] In one aspect, order tracker 230 may be configured to interact with procurement-specific data for retrieving the status of an order. Order tracker 230 is used to track an order from placement of the order to final delivery of the order using procurement-specific data sources for retrieving the status of the order.

[0056] In an example, the updater 150 may be configured such that the problem predictor 240 may deploy the order tracking data and the behavior model generated by the order tracker 230 to predict potential events related to the received order to be fulfilled using the procurement interaction based on the behavior model. The potential event may indicate a problem that affects the order. In an example, the problem predictor 240 is used to investigate previously encountered problems during any historical procurement interactions to identify problems that affect the order. The problem predictor 240 may perform an investigation of exceptions previously encountered during the sales process and problems previously generated during the sales process to identify potential events or problems that affect the order. Additionally, in an example, as part of the investigation, the problem predictor 240 may also collect evidence in the form of historical cases.

[0057] In an example, the problem affecting the order may be any of expired merchandise, warehouse issues, manufacturing issues (if the order needs to be manufactured first), billing issues, payment issues, shipping issues, etc. Billing issues may include not receiving the bill if the merchandise was received, not generating the bill if the merchandise was shipped.

[0058] In one aspect, the model deployer 260 of the problem predictor 240 can attempt to capture various potential events from the steps of the procurement operation through the behavioral model, wherein each step has been identified as a problem-causing juncture affecting an order in the procurement operation. In an example, as soon as an order is placed by a customer or received by a service provider, as the case may be, the problem predictor 240 is triggered to deploy the behavioral model.

[0059] As part of deploying the behavioral model, the model deployer 260 uses the behavioral model to predict any potential events related to orders that will be fulfilled using procurement operations. The model deployer 260 can predict potential events to indicate issues that can directly or indirectly affect orders. For example, a direct impact can be in the form of a real-world event (such as a natural disaster) that can affect order fulfillment. In another example, a direct impact can be a shipment delay due to various regulations and compliance issues. An indirect impact can be in the form of an economic, political, or social situation brewing in an area that can affect demand or supply and, in turn, orders.

[0060] Additionally, the model deployer 260 can predict the affected orders from the perspective of the party implementing the system 110 (the service provider or the customer). In either case, the model deployer 260 can identify in an example that the potential events can be the same and can affect the customer and the service provider equally. As previously mentioned, the model deployer 260 uses the behavioral model and the dates received from the order tracker 230 to track the sales operation from order placement to final delivery, thereby evaluating each failure point along the path to accurately predict the potential events, subsequently predict their impact on the orders, and can recommend possible solutions.

[0061] In an example, as part of the prediction of a potential event, the model deployer 260 may generate a hypothesis based on historical data; and provide a recommendation to fix the problem based on the hypothesis. The problem may be a problem that affects an order. If an event that may potentially affect an order is predicted by the problem predictor 240, the parser 270 may be configured to proactively fix the problem that affects the order to automatically troubleshoot the order.

[0062] For example, in one case, as part of proactive remediation, the model deployer 260 can generate hypotheses based on historical situation data, and then build recommendations based on the hypotheses for remediating the problem. The parser 270 can perform the actions recommended by the model deployer 260 based on the hypotheses for remediating the problem. In an example, the recommendations can be in the form of action items that can be performed by implementers of the system 110 to mitigate the impact of the problem on orders or avoid the impact altogether.

[0063] Additionally, as part of the automation of order troubleshooting, the updater 150 also includes a dialog manager 250. The dialog manager 250 can be configured to facilitate user interaction with procurement-specific data sources as required by issues affecting orders. Figure 7 Explain in detail the user interactions with procurement-specific data sources.

[0064] The updater 150 may be configured to generate an alert to a user via the dialogue manager 250 in response to a prediction of a potential event. For example, the system 110 may alert all stakeholders, such as the sales team, operations team, delivery team, and verification team, in response to a prediction of a potential event in a sales operation, which may allow stakeholders to preemptively provide solutions to problems associated with the predicted event. In an example, the updater 150 may be configured to generate an automated communication with at least one of a procurement-specific data source and user via the dialogue manager 250 following a prediction of a problem affecting an order. In an example, the dialogue manager 250 may generate an alert to a stakeholder regarding the completion of an order and provide for clearing of payment. The alert generator may be, for example, an automated communication, a voice message, a text message, a chat interface, and an email.

[0065] In addition, the system 110 may also provide for the aggregation of newly discovered influencing factors (which were previously unknown or not considered in the behavioral model). Therefore, in one example, the updater 150 includes a model enhancer 280. The model enhancer 280 may be configured to investigate an order based on an operational behavior pattern associated with the order to determine whether the order conforms to the behavioral model. For example, the model enhancer 280 may determine whether the current order falls within a category of a known operational behavior pattern or whether it is an anomaly. The model enhancer 280 may generate a hypothesis for the order when the order does not conform to the behavioral model. The hypothesis may indicate that the behavioral model does not accurately reflect the operational pattern. The model enhancer 280 may incorporate the hypothesis into the behavioral model to predict potential events that affect the order.

[0066] In the event that the model enhancer 280 determines that the order does not conform to the behavior model, the model enhancer 280 also generates a hypothesis for the order and incorporates the hypothesis into the behavior model to enable prediction of potential events in such an event. In an example, the model enhancer 280 can generate a recommendation for a change in the behavior model based on the hypothesis that will accommodate the impact due to the new order.

[0067] Figure 3 An overall view of an order behavior model and an environmental behavior model according to an example embodiment of the present disclosure is illustrated. In one aspect, the behavior model constructed by the behavior model builder 130 may be an order behavior model. In an example, the behavior model constructed by the behavior model builder 130 may be an environmental behavior model. Both the order behavior model and the environmental behavior model are required in the successful execution of an order. Figure 3 Depicted is the interaction of order behavior model 302 and environmental behavior model 338. Order behavior model 302 and environmental behavior model 338 may be separated by dashed line 340. Dashed line 340 is for demonstration purposes only.

[0068] The order behavior model 302 includes a purchase order 306. The purchase order 306 may be an order received by the monitor 140 through a user's purchasing interaction. In an example, the purchase order 306 may be a new order for the purchase of goods and services. In an example, the purchase order 306 may include an order for an expired item 304. The expired item 304 order may be a subsequent order resulting from the expiration of a previously placed order as depicted by box 304. At box 320, the purchase order 306 may be received by the service provider. Box 306 of the purchase order and box 320 of the received order depict the interaction of the order behavior model 302 with the environmental behavior model 338. The purchase order 306 and the received order 320 may be the same and may be used interchangeably herein and may be depicted as different boxes due to differences in the behavior models. The updater 150 will begin tracking orders at step 306 and will interchange between the order behavior model 302 and the environmental behavior model 338. For simplicity, Figure 3 The description will only mention the exemplary steps depicted for the order behavior model 302 and the environment behavior model 338.

[0069] In the example, if the purchase order 306 includes manufacturing of the goods, the received order 320 is sent to the manufacturing unit 324. The manufacturing unit 324 can send the received order 320 to the warehouse 322 after the manufacturing of the received order 320 is completed. In the example, if the purchase order 306 does not require manufacturing, it is sent to the warehouse 322. Subsequently, the received order 320 can be sent to the shipping unit 326. Figure 2The depicted order tracker 230 tracks the progress of a purchase order 306 through a received order box 320, a manufacturing unit 324, a warehouse 322, and a shipping unit 326. Herein, the environment behavior model 338 interacts with the order behavior model 302 and the purchase order 306 is depicted as a commodity received at box 308 in the order behavior model 302. Step 308 will only be completed when the commodity is received by the customer. In an example, the commodity can be ordered through the purchase order 304. In an example, the commodity can be ordered earlier and can expire, as depicted by box 304.

[0070] The updater 150 may track the expense list 328 after the item is received by the customer at 308. As depicted by block 312, the expense list 328 may be sent to the customer after the item has been received by the customer. In an example, the expense list 328 may be part of the environmental behavior model 338. In an example, the order behavior model 302 may recognize that the item has been received at block 308, however, as depicted by block 310, the expense list has not yet been received. The dialogue manager 250 of the updater 150 will then interact with the stakeholder and notify that the expense list was not received. The dialogue manager 250 may generate an alert until the expense list 328 is received, as depicted by block 312.

[0071] The updater 150 may perform a three-way match after receiving the expense list 312 at block 314. At block 314, the purchase order 306, the received goods 308, and the received expense list 312 are matched together. After a successful three-way match, at block 316, the account for the purchase order 306 becomes available for payment. In an example, as depicted by block 318, the updater 150 may facilitate payment. On the service provider's side, once the expense list 328 is generated, the account for the received order 320 becomes available for account receipt 330. In an example, if the system 110 is deployed by a service provider, the updater 150 will generate a warning for payment overdue at block 332. In an example, the dialogue manager 250 may facilitate payment collection 334. The order tracker 230 will track the payment for the purchase order 306 until the final payment is made or received at block 336. The system 110 may be deployed by clients as well as service providers and the system 110 will deploy behavioral models to identify payments to track until final payment 336 is received.

[0072] Figure 4 The order behavior model 400 according to an example embodiment of the present disclosure is illustrated. In an example, the order behavior model 400 can be used with Figure 3The order behavior model 400 is the same as the order behavior model 302 depicted. The order behavior model 400 can be related to the shipment of goods related to the order. It should be understood that there can be many similar behavior models that the system 110 can make for various orders. The system 110 can make multiple behavior models for a single order. As described above, all behavior models are stored in a library and used for enhancement, formation, and deployment of current models and future models. Any of the behavior models can be used for intelligent order processing.

[0073] According to an example of the present disclosure, an order behavior model 400 (also referred to herein as model 400) includes the step of transmitting a purchase order at block 402. The purchase order may be purchase order 306. The purchase order may be a new order different from purchase order 306. The transmitted purchase order 402 is followed by receiving a confirmation 404. In an example, if the system 110 has been deployed by the customer, the confirmation 404 may be received from the service provider's end. In an example, if the system 110 has been deployed by the service provider, the confirmation 404 may be received from the customer's end. After receiving the confirmation 404, the model 400 receives a confirmation 406 regarding the shipment of the items included in the order. In an example, there is a problem regarding the shipment of the order, and the model 400 records the same as the shipping exception 408. The order tracker 230 may then continuously query the procurement-related sources related to the order regarding the shipping information. The parser 270 will perform a fix regarding the shipping exception 408 or alert the stakeholders. The process will continue until the shipping exception 408 is resolved and the confirmation 406 is received by the order tracker 230. In an example, model enhancer 280 will record shipping exception 408 for future reference and updating of model 400. In an example, model 400 may be constructed so that problem predictor 240 may predict shipping exception 408 and resolver 270 may perform proactive remediation so that shipping exception 408 may be avoided.

[0074] In an example, when the merchandise has been received by the customer, a confirmation 412 may be received by the system 110. In an example, the merchandise may not be received by the customer, and then a receipt exception 410 is recorded by the model 400. The order tracker 230 may then continuously query the procurement-related sources associated with the order for merchandise receipt information. The parser 270 will perform a fix or alert the stakeholders regarding the receipt exception 410. The process will continue until the receipt exception 410 is resolved and the confirmation 412 is received by the order tracker 230. In an example, the model enhancer 280 will record the receipt exception 410 for future reference and updates of the model 400. In an example, the model 400 may be constructed so that the problem predictor 240 can predict the receipt exception 410 and the parser 270 can perform proactive repairs so that the receipt exception 410 can be avoided. After receiving the confirmation 412, the model 400 includes a validation 414. The validation 414 may include checking whether the received merchandise or service is correct. In an example, any of the components of the system 110 may be required in the construction and operation of the model 400.

[0075] Figure 5 FIG. 5 shows an environment behavior model 500 according to an example embodiment of the present disclosure. In an example, the environment behavior model 500 can be used with Figure 5 The order behavior model 500 (also referred to herein as model 500) may be related to the manufacture and delivery of goods related to an order. Model 500 may be considered an example of the environmental behavior model 338. It should be understood that there may be many similar behavior models that the system 110 may make for various orders. The system 110 may make multiple behavior models for a single order. As described above, all behavior models are stored in a library and used for the enhancement, formation, and deployment of current and future models. Any of the behavior models may be used for intelligent order processing. Any of the components of the system 110 may be required in the construction and operation of the model 500.

[0076] According to an example of the present disclosure, model 500 includes a step of receiving an order 502. The received order 502 may be associated with a Figure 3320. In the example, the received order 502 may be different from the received order 320. In the illustrated example, the model 500 incorporates three paths for the received order 502. The first path may be where the received order 502 is a commodity that the service provider has in stock. The second path may be where the received order 502 is a commodity that the service provider has in stock but does not have a sufficient number of units to meet the requirements of the received order 502. The third path may be where the received order may be unique and may require manufacturing in order to meet the requirements of the received order 502. The model 500 may be configured so that it will continuously query the procurement-related sources and select one of the three paths mentioned above and described herein. These three paths are presented below.

[0077] In an example, where the received order 502 is for an item that the service provider has in stock, the received order 502 may be sent to an inventory unit 504. The inventory unit 504 will provide information about whether there are a sufficient number of units in the inventory 504 to satisfy the requirements of the received order 502. If there are a sufficient number of units in the inventory 504 to satisfy the requirements of the received order 502, the received order 502 may be sent to a warehouse 506.

[0078] In an example, where the received order 502 is for a commodity that the service provider has in stock but does not have a sufficient number of units to meet the requirements of the received order 502. The model provides a shortage indication 510. The shortage indication 510 sends the received order 502 to a scheduling unit 514. The scheduling unit 514 can be configured to schedule the manufacturing of the commodity related to the received order 502. The scheduling unit 514 can send the received order 502 to a tool calibration unit 516. The tool calibration unit 516 can perform investigations on the received order 502 to calibrate the tools and machines used to manufacture the commodity in order to complete the requirements of the received order 502. After successful tool calibration, the received order 502 can be sent to the manufacturing and assembly unit 518 and confirmation is sought by the order tracker 230 regarding the complete manufacturing and assembly of the commodity related to the received order 502.

[0079] In an example, the manufacturing and assembly unit 518 is unable to meet the manufacturing requirements of the received order 502, and the model 500 records the manufacturing exception 520. In such an example, the order tracker 230 can then continuously query the procurement-related sources related to the order for manufacturing information. The parser 270 will perform a fix or alert the stakeholders regarding the manufacturing exception 520. The process will continue until the manufacturing exception 520 is resolved and confirmation from the manufacturing and assembly unit 518 is received by the order tracker 230. In an example, the model enhancer 280 will record the manufacturing exception 520 for future reference and update of the model 500. In an example, the model 500 can be constructed so that the problem predictor 240 can predict the manufacturing exception 520 and the parser 270 can perform proactive repairs so that the manufacturing exception 520 can be avoided. After confirmation from the manufacturing and assembly unit 518, the received order 502 can be sent to the warehouse 506.

[0080] In the example, the received order may be unique and may require manufacturing to meet the requirements of the received order 502. The model 500 provides an indication 512. The indication 512 may indicate that fulfilling the received order 502 may require unique processing and manufacturing. The received order is sent to a scheduling unit 514. All steps described above for the second path are repeated here, where the received order 502 is a commodity that the service provider has in stock but does not have a sufficient number of units to meet the requirements of the received order 502. These steps are not mentioned again for the sake of brevity and simplicity of the document.

[0081] All three paths described above will result in received order 502 being sent to warehouse 506. Figure 5 In the example illustrated in , received order 502 may be sent from warehouse 506 to international shipping unit 522. System 110 may receive confirmation therefor. In the example, there is a problem with the shipping of the order, and model 500 records the same as shipping exception 526. Order tracker 230 may then continuously query procurement-related sources related to the order for shipping information. Parser 270 will perform a fix or alert stakeholders regarding shipping exception 526. The process will continue until shipping exception 526 is resolved and confirmation from international shipping unit 522 is received by order tracker 230. In the example, model enhancer 280 will record shipping exception 526 for future reference and update of model 500. In the example, model 500 may be constructed so that problem predictor 240 can predict shipping exception 526 and parser 270 can perform proactive fixes so that shipping exception 526 can be avoided.

[0082] In the illustrated example, the received order 502 may be sent from the international shipping unit 522 to the domestic shipping unit 524. The system 110 may receive a confirmation about it. In the example, there is a problem with the shipping of the order, and the model 500 records the same as the shipping exception 528. The order tracker 230 may then continuously query the procurement-related sources related to the order for shipping information. The parser 270 will perform a fix or alert the stakeholders about the shipping exception 528. The process will continue until the shipping exception 528 is resolved and a confirmation from the international shipping unit 524 is received by the order tracker 230. In the example, the model enhancer 280 will record the shipping exception 528 for future reference and updates of the model 500. In the example, the model 500 may be constructed so that the problem predictor 240 can predict the shipping exception 524 and the parser 270 can perform proactive repairs so that the shipping exception 528 can be avoided. Subsequently, when the goods have been received by the customer, the confirmation 530 may be received by the system 110.

[0083] The order tracker 230 of system 110 can continuously query procurement related sources and obtain confirmation and order status from any of the following: scheduling unit 514, tool calibration unit 516 and manufacturing unit and assembly unit 518, warehouse 506, international shipping unit 522, domestic shipping unit 524 and confirmation 530 that the goods have been received.

[0084] Figure 6 A monitoring and automated investigation model 600 of orders according to an example embodiment of the present disclosure is illustrated.

[0085] As described above, the updater 150 of the system 110 is used to investigate previously encountered problems during any historical procurement interactions to identify problems that affect orders. The updater 150 may also include a model enhancer 280 for enhancement, updating, and calibration of the behavior model. Monitoring of the retrieval status of orders and investigation of previously encountered problems, investigation of orders that meet the behavior model are illustrated in the illustrated monitoring and automatic investigation model 600 (referred to herein as model 600). Model 600 may require deployment of any of the components of the system 110.

[0086] In the example, model 600 is related to the shipping of goods related to the order. As described above, order tracker 230 can interact with procurement-specific data sources to retrieve order status. In the example, shipping exception 602 can be detected while processing an order. The order can be similar to any of orders 306, 320, 402, and 502. In the example, there is a problem with the shipping of the order, and model 600 records the same content as shipping exception 602. Order tracker 230 can then continuously query procurement-related sources related to the order for shipping information. Behavior identifier 210 can monitor procurement-specific data sources associated with shipping exception 602. Behavior identifier 210 can identify operational behavior patterns from the monitored procurement-specific data related to exception 302. Modeler 220 will build a behavioral model based on the operational behavior patterns related to shipping exception 602. Problem predictor 240 can use model deployer 260 to deploy the behavioral model of shipping prediction 604. Prediction 604 can be a potential event related to the received order to be fulfilled using procurement interaction. The order tracker 230 will monitor the shipping exceptions 602 by querying various procurement specific data sources. Once recorded, the shipping exceptions 602 will be flagged for the shipping prediction 604. The system 110 is configured so that the behavior model builder 130 and updater 150 will monitor the prediction 604 continuously.

[0087] In the example, as illustrated by the model 600, the shipping prediction 604 can follow the 3 paths presented below. The first path can be that the shipping of the order is timely, the second path can be that the shipping of the order is earlier than expected, and the third path can be that the shipping of the product may be delayed. Each of these is explained in detail below. As mentioned, as depicted by box 608, the first path can be that the shipping prediction 604 will predict the shipping of the order in time. In such an example, the shipping prediction 604 is sent back to the monitoring of the shipping exception 602. As depicted by box 610, the second path can be that the shipping prediction 604 will predict the shipping of the order early. In such an example, the shipping prediction 604 is sent back to the monitoring of the shipping exception 602. As previously mentioned, as depicted by box 612, the third path can be that the shipping prediction 604 will predict the shipping of the order late. In such an example, the model enhancer 280 can generate a hypothesis 614. The hypothesis 614 can be based on a pre-existing behavioral model library 616. For example, the library 616 can include shipping delays in various segments. Assumption 614 may select a segment in which the shipping delay of block 612 may be explained. The dialog manager 250 of the system 110 may then initiate communication 618. The communication 618 may be an email, a phone call, a virtual agent, etc. Figure 7The communication 618 is explained in detail. Subsequently, the model 600 will find a cause 620 for the shipping delay based on the hypothesis 614 and the communication 618. Additionally, the model 600 may make a selection 622. The selection 622 may be a new segment that may have caused the shipping delay based on the confidence level of the hypothesis 614. The process of initiating the communication 618 described above may be repeated by the model 600. Additionally, the model 600 obtains a new estimate 624 of the shipping time. It must be understood that any of the components of the described system 110 may be used to perform the techniques described by the model 600.

[0088] In an example, the resolver 270 will perform a fix or alert stakeholders regarding the delay 612. The process will continue until the delay 612 is resolved and confirmation on the time delivery 608 is received by the order tracker 230. In an example, the model enhancer 280 will record the delay 612 for future references and updates to the model 600. In an example, the model 600 can be constructed so that the problem predictor 240 can predict the delay 612 and the resolver 270 can perform proactive fixes so that the delay 612 can be avoided.

[0089] Figure 7 An example of dialog management for remediation of issues related to fulfillment of an order according to an example embodiment of the present disclosure is illustrated. Figure 7The dialog management model 700 is illustrated, which should be taken as an example only. In the example, the dialog manager 250 can be used to perform the techniques described herein. It should be understood that any of the components of the system 110 can be used to perform the techniques described herein. The model 700 may include a multichannel dialog manager 702. In the example, the multichannel dialog manager 702 may use technologies such as natural language generation (NLG) 714 and natural language understanding (NLU) 710. NLG 714 and NLU 710 may be well known to those skilled in the art and will not be explained here for the sake of brevity of the document. The multichannel dialog manager 702 initiates an interaction with a supplier 712. The interaction may be in the form of, for example, an email 704, a chat 706, or an SMS 708. The multichannel dialog manager 702 may interact with a supplier behavior model 724. The supplier behavior model 724 may be a new model built by the behavior model builder 130. In the example, the supplier behavior model 724 may be one of the pre-existing models stored in the system 110. The interaction of the multichannel dialogue manager 702 with the supplier behavior model 724 may be facilitated by a transformation 718. The transformation 718 may include communications received from the supplier 712. In an example, the transformation 718 may be configured to decipher the transformation context 716. In an example, the transformation 718 may also include information from an action planner 720. The action planner 720 may be derived from a case activity engine 722. The action planner 720 and the case activity engine 722 may include information about various investigations performed by the updater 150 on the supplier behavior model 724. The transformation 718 may serve as an interface between the supplier behavior model 724 and the multichannel dialogue manager 702.

[0090] Figure 8 Hypothesis generation and evidence collection for remediation of problems related to fulfillment of orders according to other example embodiments of the present disclosure are illustrated. Figure 8 Case 800 is illustrated, which illustrates an investigation performed by the updater 150 through the problem predictor 240 and the model enhancer 280. As described above, the model deployer 260 of the problem predictor 240 can be configured to generate hypotheses based on historical data; and provide recommendations to fix the problem based on the hypothesis. In addition, the model enhancer 280 can be configured to investigate the order based on the operational behavior pattern associated with the order to determine whether the order conforms to the behavior model. In addition, the model enhancer 280 can generate a hypothesis for the order when the order does not conform to the behavior model. In an example, the model enhancer 280 can incorporate the hypothesis into the behavior model to predict potential events. Case 800 describes the process of hypothesis generation and evidence collection to form a basis for remediation.

[0091] Case 800 may include recording an exception event 802. Model deployer 260 may investigate and find, for example, two types of exceptions that may be possible for exception event 802. These exceptions may include a commodity reception exception 804 and a shipping exception 806. Model deployer 260 may further investigate each of commodity reception exception 804 and shipping exception 806 and obtain different hypotheses for each. Each hypothesis may then be investigated whether there is evidence in order to determine the confidence level of the hypothesis. It should be noted that the hypothesis formed and the evidence collected for commodity reception exception 804 may also be interconnected with the hypothesis formed and the evidence collected for shipping exception 806. In the example, shipping exception 806 may give rise to hypothesis 822, hypothesis 826, hypothesis 830, and hypothesis 834 (explained below). For the sake of brevity, all assumptions and evidence are explained below, and those of ordinary skill in the art will understand that each hypothesis and the accompanying evidence may be connected to each other hypothesis and the accompanying evidence.

[0092] In an example, the merchandise received exception 804 may lead to a hypothesis that there may be a delay in shipping. The model deployer 260 may further investigate and form a hypothesis 810 that references delays in international shipping, a hypothesis 814 that references delays in domestic shipping, and a hypothesis 816 that references delays due to shipping accidents or damaged merchandise. The hypothesis 810 may lead to the system 110 performing a check 812 on the shipping company responsible for international shipping of the order associated with the merchandise received exception 804. In an example, the system 110 may use the order tracker 230 and the dialogue manager 250 to perform the check 812. The check 812 may form evidence for recommending a course of action for the hypothesis 810. The hypothesis 814 may lead to the system 110 performing a check 816 on the shipping company responsible for domestic shipping of the order associated with the merchandise received exception 804. In an example, the system 110 may use the order tracker 230 and the dialogue manager 250 to perform the check 816. The check 816 may form evidence for recommending a course of action for the hypothesis 814. Assumption 818 may cause system 110 to perform a check 820 regarding the shipping company responsible for shipping the order associated with merchandise received exception 804. In an example, system 110 may use order tracker 230 and dialog manager 250 to perform check 820. Check 820 may form evidence for recommending a course of action for assumption 818.

[0093] Assumption 822 may refer to a commodity that is out of stock and awaiting its arrival. Assumption 822 may cause system 110 to perform a check 824 on a warehouse company responsible for warehousing of orders related to shipping exception 806. In an example, system 110 may use order tracker 230 and dialogue manager 280 to perform check 824. Check 824 may form evidence for recommending a course of action for assumption 822. In an example, assumption 826 may refer to a commodity that is out of stock and awaiting manufacturing. Assumption 826 may cause system 110 to perform a check 828 on a manufacturing company responsible for manufacturing of orders related to commodity received exception 804. In an example, system 110 may use order tracker 230 and dialogue manager 250 to perform check 828. Check 828 may form evidence for recommending a course of action for assumption 826. In one aspect, assumption 830 may refer to a commodity that is out of stock and can no longer be produced. Assumption 830 may cause system 110 to perform check 832 regarding management of a product manufacturing company responsible for manufacturing of an order associated with merchandise received exception 804. In an example, system 110 may use order tracker 230 and dialogue manager 250 to perform check 832. Check 832 may form evidence for recommending a course of action for assumption 830. In an example, assumption 834 may refer to shipping being unavailable. Assumption 834 may cause system 110 to perform check 836 regarding a shipping company responsible for shipping of an order associated with merchandise received exception 804. In an example, system 110 may use order tracker 230 and dialogue manager 250 to perform check 836. Check 836 may form evidence for recommending a course of action for assumption 834.

[0094] In the example, all of the checks 812, 816, 820, 824, 828, 832, 836 mentioned above can lead the system 110 to a conclusion 838. The conclusion 838 may include a root cause analysis of the hypothesis and supporting evidence. The model deployer 260 can then deploy the behavior model to predict potential events related to the received order to be fulfilled using the procurement interaction based on the behavior model, where the potential event indicates a problem that affects the order. The parser 270 can proactively fix the problem that affects the order to automatically fix the order. The model enhancer 280 can incorporate each of the assumptions and supporting evidence into the behavior model to predict the potential event. It should be noted that any of the components in the system 110 can be used to perform the technology presented by the case 800.

[0095] Fig. 9 The present invention is based on an exemplary embodiment of the present disclosure. Figure 4 Example of next order prediction using the order behavior model. Fig. 9An order behavior model 900 (also referred to herein as model 900) is illustrated, which shows an example of an order behavior model 900 based on Figure 4 In an example, model 900 may be different from order behavior model 400.

[0096] The order behavior model 900 may be related to the delivery of goods related to the order. It should be understood that there may be many similar behavior models that the system 110 may make for various orders. The system 110 may make multiple behavior models for a single order. As described above, all behavior models are stored in a library and used for enhancement, formation, and deployment of current and future models. Any behavior model may be used for intelligent order processing.

[0097] According to an example of the present disclosure, the order behavior model 900 includes the step of transmitting a purchase order at box 902. The purchase order can be purchase order 306, purchase order 402. The purchase order can be a new order different from purchase order 306 or purchase order 402. The transmitted purchase order 902 is followed by a receiving confirmation 904. In an example, if the system 110 has been deployed by the customer, the confirmation 904 can be received from the service provider's side. In an example, if the system 110 has been deployed by the service provider, the confirmation 904 can be received from the customer's side. The problem predictor 240 can be configured to make a prediction 906 for receiving the confirmation 904.

[0098] After receiving confirmation 904, model 900 receives confirmation 914 regarding the shipment of the merchandise included in the order from the warehouse. In the example, there is a problem with the shipment of the order, and model 900 records the same as shipping exception 908. Order tracker 230 can then continuously query procurement-related sources related to the order for shipping information. Parser 270 will perform a fix or alert stakeholders regarding shipping exception 908. The process will continue until shipping exception 908 is resolved and confirmation 914 is received by order tracker 230. In the example, model enhancer 280 will record shipping exception 908 for future reference and update of model 900. In the example, model 900 can be constructed so that problem predictor 240 can make prediction 912. Prediction 912 can refer to the potential event of shipping exception 908 occurring, and parser 270 can perform proactive repair so that shipping exception 908 can be avoided. Similarly, system 110 can be configured to make prediction 910 for receiving confirmation 914.

[0099] In an example, when the merchandise has been received by the customer, a confirmation 920 may be received by the system 110. In an example, the merchandise may not be received by the customer, and then a receipt exception 918 is recorded by the model 900. The receipt exception 918 may also be recorded in the event that an incorrect merchandise may be received by the customer. The order tracker 230 may then continuously query the procurement-related sources associated with the order regarding merchandise receipt information. The parser 270 will perform a fix or alert the stakeholders regarding the receipt exception 918. The process will continue until the receipt exception 918 is resolved and the confirmation 920 is received by the order tracker 230. In an example, the model enhancer 280 will record the receipt exception 918 for future reference and updates of the model 900. In an example, the model 900 may be constructed so that the problem predictor 240 may make a prediction 916. The prediction 916 may refer to a potential event of the receipt exception 918, and the parser 270 may perform a proactive fix so that the receipt exception 918 may be avoided. Similarly, the system 110 may be configured to make a prediction 926 for the receipt confirmation 920. Prediction 926 may refer to the occurrence of a potential event for receiving exception 918 .

[0100] After receiving confirmation 920, model 900 includes verification 924. Verification 924 may include checking whether the received goods or services are correct. In an example, model 900 may be constructed so that problem predictor 240 can make predictions 922. Prediction 922 may refer to a potential event of verification 924 not working as required. Parser 270 may perform proactive repairs so that potential events of verification 924 not working as required can be avoided. In an example, any of the components of system 110 may be required in the construction and operation of model 900.

[0101] Fig.10 An example of a fix for a problem related to fulfillment of an order according to an example embodiment of the present disclosure is illustrated. Fig.10 An example case 1000 is illustrated relating to a root cause determination for an automated investigation performed by the system 110 as also described above. Case 1000 may be a subsequent case to case 800, which illustrates an investigation performed by the updater 150 through the problem predictor 240 and the model enhancer 280. Case 1000 may include a root cause determination 1002. Root cause determination 1002 may be the result of an investigation performed by the updater 150 through the problem predictor 240 and the model enhancer 280. As described above through Figure 8As mentioned, all checks of case 800 - 812, 816, 820, 824, 828, 832, 836 - can lead the system 110 to a conclusion 838. Conclusion 838 can include a hypothesized root cause analysis and supporting evidence. Conclusion 838 can be the same as root cause determination 1002. Root cause determination 1002 can be different from conclusion 838. Root cause determination 1002 can lead to a determination of a remediation action. In an example, parser 270 can be configured to proactively remediate issues affecting an order to automatically remediate the order. It should be understood that any of the components of system 110 can be used for the remediation described herein.

[0102] In the example case 1000, root cause determination 1002 results in the determination of various issues that may be potential events that affect orders. Each possibility may result in a different remediation method that may be deployed by the parser 270 or system 110 for proactive remediation of the issue. The following explains various issues that may be identified through the case 1000 and the accompanying remediation actions.

[0103] In the example, the root cause determination 1002 leads to a problem 1004 regarding delays in international shipping. The parser 270 may then perform identification 1018. The identification 1018 may include a workaround method for fixing the problem 1004. The parser 270 may deploy a method 1032 as part of the fix for the problem 1004. The method 1032 may include using an alternative product or shipper to avoid the problem 1004, which refers to delays in international shipping. In the example, the root cause determination 1002 leads to a problem 1006 regarding delays in domestic shipping. The parser 270 may then perform identification 1020. The identification 1020 may include a workaround method for fixing the problem 1006. The parser 270 may deploy a method 1034 as part of the fix for the problem 1006. The method 1034 may include using an alternative product or shipper to avoid the problem 1006, which refers to delays in domestic shipping. In the example, the root cause determination 1002 leads to a problem 1008 regarding a shipping incident. Parser 270 may then perform identification 1022. Identification 1018 may include finding an alternative shipper as a workaround method for fixing problem 1008. Parser 270 may deploy the method found by identification 1022 as part of the fix for problem 1008.

[0104] In an example, the root cause determination 1002 results in a problem 1010 regarding a warehouse problem. The parser 270 may then perform an identification 1024. The identification 1024 may include finding that the warehouse can ship items to inventory that fixes the problem 1010. The parser 270 may deploy a method 1036 as part of the fix for the problem 1010. The method 1036 may include using an alternative product or shipper to avoid the problem 1010, which refers to a warehouse problem. In an example, the root cause determination 1002 results in a problem 1012 regarding a manufacturing problem. The parser 270 may then perform an identification 1026. The identification 1026 may include finding a contingency design for fixing the problem 1012. The parser 270 may deploy a method 1038 as part of the fix for the problem 1012. The method 1038 may include ensuring that the new design is rated according to the order requirements and can be used.

[0105] In the example, the root cause determination 1002 leads to the problem 1014 regarding the inventory problem. The parser 270 may then perform identification 1028. If waiting for inventory is not an option for fixing the problem 1014, then the identification 1028 may include finding a substitute. The parser 270 may deploy a method 1040 as part of the fix for the problem 1014. The method 1040 may include ensuring that the substitute is rated according to the order requirements and can be used. In the example, the root cause determination 1002 leads to the problem 1016 regarding the inventory problem. The parser 270 may then perform identification 1030. The identification 1028 may include finding a substitute or alternative product for fixing the problem 1016. The parser 270 may deploy a method 1042 as part of the fix for the problem 1016. The method 1042 may include ensuring that the substitute is rated according to the order requirements and can be used.

[0106] Fig.11 An example of ongoing inventory maintenance resulting from deploying a smart order concierge according to an example embodiment of the present disclosure is illustrated. Fig.11A case 1100 is illustrated in which a customer or service provider may be able to maintain a near constant state of inventory according to the deployment of the system 110. The case 1100 may be described using three inventory segments, namely an incoming segment 1102, an inventory segment 1104, and an outgoing segment 1106. The incoming segment 1102 may refer to merchandise that may be purchased for the inventory segment 1104. The outgoing segment 1106 may refer to merchandise that may leave the inventory segment 604 to be used elsewhere or shipped out as required. The case 1100 may deploy the system 110 for maintaining a constant state of the inventory segment 1104. For example, when an increase in the incoming segment 1102 may be detected, the system 110 may help schedule a delivery date at an earlier time than otherwise would have been scheduled. When an increase in the outgoing segment 1106 may be detected, the system 110 may help schedule a delivery date at a later time than otherwise would have been scheduled. Wherein a constant level of the inventory segment 1104 may be ensured.

[0107] Fig.12 A hardware platform 1200 for implementing the system 110 according to an embodiment of the present disclosure is illustrated. In particular, computing machines such as, but not limited to, internal / external service clusters, quantum computers, desktops, laptops, smart phones, tablets, and wearable devices may be used to execute the system 110 or may have the structure of the hardware platform 1200. The hardware platform 1200 may include additional components not shown and some of the components described may be removed and / or modified. In another example, a computer system with multiple GPUs may be located on an external cloud platform including Amazon Web Services or an internal company cloud computing cluster or organizational computing resources, etc.

[0108] refer to Fig.12 , hardware platform 1200 may be a computer system 1200 that may be used with the examples described herein. Computer system 1200 may represent a computing platform including components that may be in a server or another computer system. Computer system 1200 may be executed by a processor (e.g., a single or multiple processors) or other hardware processing circuits to perform the methods, functions, and other processes described herein. These methods, functions, and other processes may be embodied as machine-readable instructions stored on a computer-readable medium, which may be non-transitory, such as a hardware storage device (e.g., RAM (random access memory), ROM (read-only memory), EPROM (erasable programmable ROM), EEPROM (electrically erasable programmable ROM), a hard drive, and flash memory). Computer system 110 may include a processor 1205 that executes software instructions or code stored on a non-transitory computer-readable storage medium 1210 to perform the methods of the present disclosure. The software code includes, for example, executing the previously referenced Figures 1 to 11Instructions of the steps described in the components of the system 110 in the example. In an example, the processor 120, the behavior model builder 130, the monitor 140, and the updater 150 can be software codes or components that perform these steps. In an example, the behavior identifier 210, the modeler 220, the order tracker 230, the problem predictor 240, the model deployer 260, the parser 270, the dialog manager 250, and the model enhancer 280 can be software codes or components that perform these steps.

[0109] The instructions on the computer readable storage medium 1210 are read and stored in the storage device 1215 or in the random access memory (RAM) 1220. The storage device 1215 provides a large space for holding static data, where at least some instructions can be stored for later execution. The stored instructions can be further compiled to generate other representations of the instructions and dynamically stored in the RAM 1220. The processor 1205 reads the instructions from the RAM 1220 and performs the actions as indicated.

[0110] Computer system 1200 also includes output device 1225 to provide at least some of the results of execution to users such as external agents as output, including but not limited to visual information. Output devices can include displays and virtual reality glasses on computing devices. For example, the display can be a mobile phone screen or a laptop screen. GUI and / or text are presented on the display screen as output. Computer system 1200 also includes input device 1030 to provide a mechanism for inputting data and / or otherwise interacting with computer system 1200 to a user or another device. The input device can include, for example, a keyboard, a keypad, a mouse or a touch screen. In the example, the output of any component of system 110 is displayed on output device 1225. Each of these output devices 1225 and input device 1230 can be connected by one or more additional peripheral devices.

[0111] A network communicator 1235 may be provided to connect the computer system 1200 to a network and in turn to other devices connected to the network, including, for example, other clients, servers, data repositories, and interfaces. The network communication 1235 may include, for example, a network adapter, such as a LAN adapter or a wireless adapter. The computer system 1200 includes a data source interface 1240 that accesses a data source 1245. A data source is an information resource. As an example, a database of exceptions and inference rules may be a data source. In addition, a knowledge repository and selected data may be other examples of a data source.

[0112] Fig.13 A method 1300 for smart order and purchase order concierge described by system 110 is illustrated according to an example embodiment of the present disclosure.

[0113] It should be understood that the method steps are shown herein for reference only, and other combinations of steps may be possible. In addition, the method 1300 may include Fig.13 For brevity, Fig.13 The description does not explain in detail Figure 1 , Figure 2 , Figure 3 , Figure 4 , Figure 5 , Figure 6 , Figure 7 , Figure 8 , Fig. 9 , Fig.10 , Fig.11 and Fig.12 The construction and operating features of the system 110 are explained in detail in the description of . The method 1300 may be performed by components of the system 110 (such as the processor 120, the behavior model builder 130, the monitor 140, and the updater 150).

[0114] The method 1300 for automated order troubleshooting can be implemented in three stages. In the first stage, the behavioral model builder 130 builds a behavioral model that captures operational behavior patterns that indicate and are associated with a process, organization, and industry, or a combination thereof, associated with a party implementing the system 110 and procurement-specific behaviors. In the second stage, the monitor 140 receives procurement interactions from a user, the procurement interactions being used to indicate queries sent by the user to the system for processing an order. The monitor 140 can facilitate the placement of orders associated with at least one of the processes, organizations, and industries related to procurement operations. Then, in the third stage, as the case may be, the updater 150 deploys the behavioral model when an order is placed or received. In the example, the updater 150 deploys the behavioral model to predict the behavior associated with the received order to be used through Figure 1 In an example, the updater 150 deploys a behavioral model to assess whether a problem affecting an order may be related to the manufacturing of the order if the order requires the manufacturing of a product before delivery. In an example, the updater 150 deploys a behavioral model to assess whether a problem affecting an order may be related to the shipping of the order after the order has left the service provider site and is in transit to the customer site. Therefore, the updater 150 can initiate an automatic fix for the problem to mitigate or completely prevent the order from being affected.

[0115] With respect to method 1300, at block 1302, procurement-specific data sources associated with processes, organizations, industries, or combinations thereof related to sales operations may be monitored. In an example, processes, organizations, and industries related to sales operations may be selected based on the operation of a party implementing the techniques of the present disclosure. Monitoring of procurement operations may be accomplished in real time by querying various databases and repositories of information online and offline from which relevant data may be retrieved. For example, social media, online portals, and other websites carrying information about processes, organizations, industries, or combinations thereof may be crawled to retrieve and monitor the above-mentioned information.

[0116] Additionally, at block 1304, operational behavior patterns may be identified based on the procurement-specific data.

[0117] Subsequently, at block 1306, a behavior model for the order is constructed using the operational behavior pattern and using a pre-existing behavior model library. In an example, the pre-existing behavior model library can be a basis for constructing the behavior model. For example, the pre-existing behavior model library can be used as a template for generating a behavior model that attempts to mimic the operational behavior pattern.

[0118] At block 1308 , the order is associated with at least one of the procurement-specific data sources.

[0119] At block 1310, a purchase interaction is received from a user. The purchase interaction may indicate a query sent by a user to the system for processing an order, which is evaluated in light of the behavioral model.

[0120] At block 1312, placement of an order associated with at least one of a process, an organization, and an industry related to procurement operations may be facilitated.

[0121] At block 1314, interaction with a procurement-specific data source may be facilitated to retrieve the status of the order. In an example, interaction with the procurement-specific data source includes tracking the order from placement of the order to final delivery of the order using the procurement-specific data source to retrieve the status of the order.

[0122] At block 1316, any potential events related to the received order to be fulfilled using procurement operations are predicted based on the behavioral model. The potential events indicate issues that may directly or indirectly affect the order. In an example, predicting the potential events includes investigating previously encountered exceptions during previously encountered issues during any historical procurement interactions to identify issues that affect the order. In an example, predicting includes investigating the order based on an operational behavior pattern associated with the order to determine whether the order conforms to the behavioral model. In an example, predicting includes generating a hypothesis for the order when the order does not conform to the behavioral model. In an example, predicting includes incorporating the hypothesis into the behavioral model to predict the potential event.

[0123] At block 1318, problems affecting the order may be proactively fixed to automatically troubleshoot the order. In an example, the fix may include generating a hypothesis based on historical data. In an example, the fix may include providing a recommendation based on the hypothesis to fix the problem.

[0124] At block 1320 , user interaction with procurement-specific data sources may be facilitated as required by issues affecting the order.

[0125] Additionally, the method 1300 may include generating an alert to a user in response to the prediction of a potential event.In an example, the method 1300 may include generating an automated communication with at least one of a procurement-specific data source and a user following the prediction of a problem affecting an order.

[0126] What has been described and illustrated herein are examples of the present disclosure and some of its variations. The terms, descriptions, and drawings used herein are set forth by way of illustration only and are not intended to be limiting. Many variations within the spirit and scope of the subject matter are possible, which are intended to be defined by the appended claims and their equivalents, wherein all terms are intended to be in their broadest reasonable sense unless otherwise indicated.

Claims

1. A system, include: processor; a memory storing instructions that, when executed by the processor, cause the system to: monitoring a procurement-specific data source associated with at least one of a process, an organization, and an industry related to procurement operations, wherein the monitoring is performed by continuously querying the procurement-specific data source to retrieve procurement-specific data; identifying an operational behavior pattern from the procurement-specific data based on predefined rules, wherein the identified operational behavior pattern indicates a current trend, a future trend, and a difference between the current trend and a previous trend for the procurement-specific data, wherein the operational behavior pattern further indicates procurement-specific data, wherein the operational behavior pattern further indicates procurement-specific behavior associated with the at least one of the process, the organization, and the industry related to procurement operations; constructing a behavior model for an order by capturing the operational behavior pattern and using a pre-existing behavior model library, wherein the pre-existing behavior model library is used as a template for generating the behavior model, the template attempting to mimic the operational behavior pattern, wherein the behavior model is any one or a combination of an order behavior model and an environment behavior model; associating the order with at least one of the purchase-specific data sources; receiving a query sent by a user to the system for processing the order; facilitating placement of the order associated with at least one of a process, an organization, and an industry related to procurement operations based on the behavioral model and the received query; retrieving a status of said order by interacting with said procurement-specific data source to track data associated with said procurement operations from said placing of said order to final delivery of said order; determining whether the order conforms to the behavior model by determining whether the order falls within a category of known behavior patterns based on the operational behavior pattern associated with the order; When the order does not conform to the behavior model: generating a hypothesis for the order to generate a new behavior pattern, the hypothesis indicating that the behavior model cannot accurately reflect the operational behavior pattern, wherein The assumptions include forecasting shipping delays in segments of the procurement operation; performing changes in the behavior model to predict potential events by incorporating the new behavior pattern into the behavior model; deploying the tracking data and the behavioral model to predict the potential events related to the received order to be fulfilled using the procurement interactions based on the behavioral model after executing the changes in the behavioral model, wherein the potential events indicate problems affecting the order, and wherein the potential events capture potential failures in the procurement operations; generating, subsequent to said prediction of said potential event, an automated communication with at least one of said procurement-specific data source and said user via at least one of email, chat, or SMS by utilizing at least one of natural language generation (NLG) techniques and natural language understanding (NLU) techniques, wherein said automated communication is facilitated by deciphering a conversation context based on information received from said user; selecting a shipping segment among the shipping segments of the procurement operation based on the confidence level of the hypothesis, and communicating with the selected shipping segment; as well as Proactively fixing the problem affecting the order based on the predicted event by identifying an automated workaround for fixing the problem, the automated workaround including deploying an alternative product, shipper, workaround design, or inventory substitution.

2. The system of claim 1, wherein the processor is configured to: Generate hypotheses based on historical data; and A recommendation is provided based on the hypothesis to fix the problem.

3. The system of claim 1, wherein the processor is operative to investigate previously encountered problems during any of historical purchasing interactions to identify the problem affecting the order.

4. The system of claim 1, wherein the processor is configured to generate a warning to the user in response to the prediction of the potential event.

5. The system of claim 1, wherein the processor is to generate an automated communication with at least one of the user and the procurement-specific data source following the prediction of the problem affecting the order.

6. A method, include: monitoring a procurement-specific data source associated with at least one of a process, an organization, and an industry related to procurement operations, wherein the monitoring is performed by continuously querying the procurement-specific data source to retrieve procurement-specific data; identifying an operational behavior pattern from the monitored procurement-specific data based on predefined rules, wherein the identified operational behavior pattern indicates a current trend, a future trend, and a difference between the current trend and a previous trend for the procurement-specific data, wherein the operational behavior pattern further indicates procurement-specific data, wherein the operational behavior pattern further indicates procurement-specific behavior associated with the at least one of the process, the organization, and the industry related to procurement operations; constructing a behavior model of the order by capturing the operational behavior pattern and using a pre-existing behavior model library, wherein the pre-existing behavior model library is used as a template for generating the behavior model, the template attempting to mimic the operational behavior pattern, wherein the behavior model is any one or a combination of an order behavior model and an environment behavior model; associating the order with at least one of the purchase-specific data sources; receiving a query sent by a user to the system for processing said order; facilitating placement of the order associated with at least one of a process, an organization, and an industry related to procurement operations based on the behavioral model and the received query; retrieving a status of said order by interacting with said procurement-specific data source to track data associated with said procurement operations from said placing of said order to final delivery of said order; determining whether the order conforms to the behavior model by determining whether the order falls within a category of known behavior patterns based on the operational behavior pattern associated with the order; When the order does not conform to the behavior model, generating a hypothesis for the order to generate a new behavior pattern, the hypothesis indicating that the behavior model does not accurately reflect the operational behavior pattern, wherein the hypothesis includes predicting a shipping delay in a segment of a procurement operation; performing changes in the behavior model to predict potential events by incorporating the new behavior pattern into the behavior model; deploying the tracking data and the behavioral model to predict the potential events related to the received order to be fulfilled using the procurement interactions based on the behavioral model after executing the changes in the behavioral model, wherein the potential events indicate problems affecting the order, and wherein the potential events capture potential failures in the procurement operations; generating, subsequent to said prediction of said potential event, an automated communication with at least one of said procurement-specific data source and said user via at least one of email, chat, or SMS by utilizing at least one of natural language generation (NLG) techniques and natural language understanding (NLU) techniques, wherein said automated communication is facilitated by deciphering a conversation context based on information received from said user; selecting a shipping segment among the shipping segments of the procurement operation based on the confidence level of the hypothesis, and communicating with the selected shipping segment; as well as Proactively fixing the problem affecting the order based on the predicted event by identifying an automated workaround for fixing the problem, the automated workaround including deploying an alternative product, shipper, workaround design, or inventory substitution.

7. The method according to claim 6, wherein the repair include: Generate hypotheses based on historical data; as well as A recommendation is provided based on the hypothesis to fix the problem.

8. The method according to claim 6, wherein said predicting said potential event include: The previously encountered exceptions during the previously encountered problem during any of the historical purchasing interactions are investigated to identify the problem affecting the order.

9. The method of claim 6, further comprising generating an alert to the user in response to the prediction of the potential event.

10. The method according to claim 6, further comprising: include: An automated communication is generated with at least one of the user and the procurement-specific data source following the prediction of the problem affecting the order.

11. A non-transitory computer-readable medium comprising machine-readable instructions executable by a processor to: monitoring a procurement-specific data source associated with at least one of a process, an organization, and an industry related to procurement operations, wherein the monitoring is performed by continuously querying the procurement-specific data source to retrieve procurement-specific data; identifying an operational behavior pattern from the procurement-specific data based on predefined rules, wherein the identified operational behavior pattern indicates a current trend, a future trend, and a difference between the current trend and a previous trend for the procurement-specific data, wherein the operational behavior pattern further indicates procurement-specific data, wherein the operational behavior pattern further indicates procurement-specific behavior associated with the at least one of the process, the organization, and the industry related to procurement operations; constructing a behavior model of the order by capturing the operational behavior pattern and using a pre-existing behavior model library, wherein the pre-existing behavior model library is used as a template for generating the behavior model, the template attempting to mimic the operational behavior pattern, wherein the behavior model is any one or a combination of an order behavior model and an environment behavior model; associating the order with at least one of the purchase-specific data sources; receiving a query sent by a user to the system for processing said order; facilitating placement of the order associated with at least one of a process, an organization, and an industry related to procurement operations based on the behavioral model and the received query; retrieving a status of said order by interacting with said procurement-specific data source to track data associated with said procurement operations from said placing of said order to final delivery of said order; determining whether the order conforms to the behavior model by determining whether the order falls within a category of known behavior patterns based on the operational behavior pattern associated with the order; When the order does not conform to the behavior model, generating a hypothesis for the order to generate a new behavior pattern, the hypothesis indicating that the behavior model does not accurately reflect the operational behavior pattern, wherein the hypothesis includes predicting a shipping delay in a segment of a procurement operation; performing changes in the behavior model to predict potential events by incorporating the new behavior pattern into the behavior model; deploying the tracking data and the behavioral model to predict the potential events related to the received order to be fulfilled using the procurement interactions based on the behavioral model after executing the changes in the behavioral model, wherein the potential events indicate problems affecting the order, and wherein the potential events capture potential failures in the procurement operations; generating, subsequent to said prediction of said potential event, an automated communication with at least one of said procurement-specific data source and said user via at least one of email, chat, or SMS by utilizing at least one of natural language generation (NLG) techniques and natural language understanding (NLU) techniques, wherein said automated communication is facilitated by deciphering a conversation context based on information received from said user; selecting a shipping segment among the shipping segments of the procurement operation based on the confidence level of the hypothesis, and communicating with the selected shipping segment; as well as Proactively fixing the problem affecting the order based on the predicted event by identifying an automated workaround for fixing the problem, the automated workaround including deploying an alternative product, shipper, workaround design, or inventory substitution.

12. The non-transitory computer readable medium of claim 11, wherein the processor is configured to: Generate hypotheses based on historical data; and A recommendation is provided based on the hypothesis to fix the problem.

13. The non-transitory computer-readable medium of claim 11, wherein the processor is to generate a warning to the user in response to the prediction of the potential event.

14. The non-transitory computer-readable medium of claim 11, wherein the processor is to generate an automated communication with at least one of the user and the procurement-specific data source following the prediction of the problem affecting the order.