Classifications

• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
  • G06Q10/00—Administration; Management
  • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
  • G06Q10/00—Administration; Management
  • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling

Definitions

• the invention relates to the processing of documents relating to items dispatched.
• An object of the invention is to facilitate the resolution of possible delivery problems.
• the invention makes it possible to determine in a very short time whether the information indicated in the document and concerning the package correspond to a situation already encountered before by the means, and if necessary to execute without delay the right action in the face of this situation.
• These operations can be performed en masse on a large number of documents by executing the action adapted to each of the documents.
• the method makes it possible to overcome the risk of error inherent in the processing of documents by people and resulting from the repetitive nature of the task.
• a person can certainly have in mind or at his disposal a number of typical situations among which it can determine that which is similar to the document in study. Whether it is stored situations or materialized documents available to the person, the comparison of the document submitted for analysis with one of these situations is particularly tedious. In addition, if the person can refer to a few dozen models, it becomes materially much more difficult, if not impossible, for her to refer to several hundred or several thousand models. On the contrary, the method of the invention makes it easy to confront the document with a very large number of models. It can therefore provide an adapted response to a wide variety of situations.
• the method of the invention not only saves time by automatically executing its steps but reduces the risks. errors and makes it possible to bring the document closer to a very large number of situations.
• the means perform a step of comparing the document with at least two predetermined document types and, when a predetermined condition on the types of document is fulfilled, a step of associating the document with one of the types.
• the automated means detect the type of document to be processed and can therefore immediately apply processing steps adapted to this type of document for better efficiency.
• This embodiment also has the advantage of dispensing with the need to sort the documents given in processing by automated means. Indeed, we can give them to process documents of different types that are not classified by types, the automated means detecting for each document to which type it belongs.
• the string preferably comprising several sentences
• the means perform an elimination of the word or group of the string prior to the calculation of the rates.
• the automated means compares the string with words or groups of words of a predetermined type recorded in the database and, if one of the words or groups of words is found in the chain, the latter is removed from the chain.
• the words or groups of words are of a predetermined type in the database, it is understood, for example, that they are associated with a specific category of data, only the data of this category being taken into account for this step. .
• the means calculate the rate only when the string does not include at least one word or a group of words of a predetermined type said prohibited in connection with the model.
• the means when the chain comprises at least one word or a group of words of a predetermined type said preferred, perform an increase of at least one of the rates prior to the determination step.
• This embodiment thus makes it possible to attribute a bonus of points if such or such word or group of privileged words is present in the chain. For example, words such as invoice, delay, etc. If such a word is detected, it is possible, for example, to increase the rate associated with each of the models of the base that includes this word. It is thus considered that the presence of such a word constitutes an element of characteristic or particularly strong resemblance between the chain and the model considered.
• the search for the preferred words can be done for example in the same way as the search for irrelevant words as explained above.
• condition or one of the conditions is that the rate exceeds a predetermined threshold.
• the automated means therefore determine whether at least one of the rates exceeds a threshold forming a minimum. This means that if a match rate is below this minimum, it is considered too low to allow for meaningful matching to the model in question. If all calculated rates are below this threshold, it can be expected that the processing of the chain by the automated means stops and that the intervention of a person is necessary to determine the sequence of operations on this chain. If one or more of the calculated rates are higher than this threshold, it is only this one or these which are taken into account for the continuation of the treatment by the automated means.
• condition or one of the conditions is that the rate is higher than the other rates.
• the model of the base that is chosen to associate it with the chain is that which corresponds to the highest rate obtained. It is therefore the one to which the chain looks the most.
• the two aforementioned conditions are combined. We can therefore predict that it is only among the rates having exceeded the threshold that the selection of the closest model is made. One can also predict that if none of the rates has exceeded the threshold, no model is selected.
• the models of the base comprising main models and secondary models respectively attached to only one of the main models and each comprising only part of the characters of the main model to which they are attached, when it is a correspondence rate between the string and one of the secondary models that fulfills the condition, the association implements the association of the article reference to the main model to which the secondary model is attached.
• the secondary models are those that are particularly similar to the corresponding main model.
• the automated means determine that the most similar model is a secondary model, it associates the corresponding main model with the chain in study.
• This embodiment makes it possible to limit the failures in the identification of a model resembling the chain. It also helps to capitalize on past experiences. Indeed, if a certain type of string is frequently encountered and considered to resemble a main model, it can be registered as a secondary model. Under these conditions, when this type of string is encountered again, it has a particularly high correspondence rate with the newly registered secondary model. It can therefore be associated with the main model with a particularly low risk of errors.
• the means record the chain in the database as a model, preferably as a main model or as a secondary model.
• the chain can be registered in the database as a main model or secondary model.
• this recording is advantageous when the string in question is found frequently.
• a computer program which includes code instructions able to control the execution of a method according to the invention when it is executed on a computer.
• the invention also provides a data recording medium comprising such a program in recorded form.
• a document processing device which comprises automated means capable of:
• a database relating to items shipped which comprises a computer medium comprising in registered form:
• an assembly which comprises a device according to the invention and a database according to the invention.
• the invention can be implemented as part of a method for inspecting dispatched articles in which:
• an automated system conjointly analyzing data of a sender of articles, relating to articles, and data of a carrier of articles, relating to articles, and
• the analysis makes it possible to compare the data of the sender and the carrier to detect early an anomaly. Most often, this early detection occurs long before the buyer has been able to detect any problem himself. In addition, it can trigger an early investigation of the carrier. The latter can then without delay examine the anomaly in question and attempt to remedy it. In such circumstances, the earlier the anomaly is detected, the easier it is to find a solution. In particular, if the item is misplaced, has followed a wrong route, or is waiting in a given location for an unexplained reason, it is all the easier to remedy if the problem is detected early. In some cases, it will even be possible to solve the problem before the buyer detects an anomaly such as a delay.
• the system receives and / or imports the data of the sender and / or the data of the carrier.
• the system determines if at least some of the data conform to at least one predetermined criterion, and if not, reports them as such.
• this preliminary control phase reduces the risks of taking into account erroneous data, which improves the reliability of the analysis carried out afterwards.
• the system renews the analysis periodically, for example at least once a day.
• the analysis is performed to determine if at least one of the items has been delivered or not.
• the method makes it possible to continue the analysis according to one or the other case in different directions.
• the analysis is performed to determine if, for each delivered item, a delivery time exceeds a predetermined threshold.
• the analysis is performed to determine if at least one of the items is awaiting delivery for a period exceeding a threshold predetermined.
• this analysis makes it possible in particular to detect if an article remains pending in a given place for an abnormally long period.
• the analysis is performed to determine whether an amount billed by the carrier exceeds a predetermined threshold.
• the action comprises preparing and / or sending a message to the carrier, the message comprising for example data from the sender and the carrier relating to at least one of the items.
• the method implements the automated preparation of a message such as a claim message.
• the sender does not have to prepare this claim.
• the system makes available to the sender of the article means for validating or even triggering the sending of the message to the carrier.
• the system analyzes a response from the carrier to the claim and preferably extracts a text from the response.
• the system makes available from a telecommunication network at least one document, such as a web page, comprising data of the sender and the carrier relating to at least one of the items, preferably the system making available several documents each relating to only one of the respective articles.
• a document such as a web page
• this document accessible to the sender and / or the buyer, allows him at any time to read the data relating to the article such as data on the state of routing, the possible existence a complaint to the carrier, or even the reply sent by the carrier, etc.
• a tracking control system for dispatched articles which comprises means capable of: - conjointly analyzing data of a sender of articles, relating to articles, and data of a carrier of articles, relating to articles, and
• a web page relating to at least one item dispatched which comprises at least the following data jointly organized in the page:
• At least one data carrier of the article relating to the article.
• FIG. 1 is a diagram of the various stakeholders involved in the implementation of the method of the invention.
• FIG. 2 is a flow chart showing the progress of the method of the invention for identifying an excessive delivery delay
• Fig. 3 is a part of the flowchart of Fig. 2 showing other steps;
• FIG. 4 is a flowchart similar to FIG. 2 showing, in particular, the processing of the carrier's responses to the complaints and the putting on line of a web page in the method of the invention;
• FIG. 5 is a view of a computer displaying this web page
• FIG. 6 is a flowchart similar to FIG. 2 illustrating the progress of the method for detecting a carrier billing anomaly
• FIG. 7 is a diagram illustrating the implementation of the method during the processing of a document and a string of characters.
• FIG. 8 is a flowchart illustrating the processing of documents in the process.
• the method according to the invention in this embodiment is used for monitoring the routing of items such as packages or folds. It is mainly implemented by an entity such as a company or an organization that we will call here a controller 2. It also involves a vendor 4 constituted, for example, also by a company or an organization, as well as a carrier. 6.
• the controller 2 comprises a system 8 comprising one or more devices 10 such as computers forming for example servers and one or more databases 12 connected to the computers 10 to be able to communicate with the latter.
• the seller 4 has a similar installation comprising one or more computers 14 and one or more databases 16 as well as the carrier 6 has one or more computers 18 and one or more databases 20.
• These organs comprise computer means such as microprocessors, memories, clocks, etc. allowing them to function as indicated below. Stakeholders 2, 4 and 6, and therefore these organs, are connected to each other by means of a telecommunication network 22, for example of the TCP / IP type, such as the Internet network.
• the seller 4 is here customer of the controller 2 for the service relating to the process.
• each of the servers comprises in registered form one or more programs comprising code instructions able to control the execution of one or more of the steps of the method when it is executed on the server.
• These programs can be recorded on a data recording medium such as a hard disk, flash memory, etc.
• the same programs can also be made available, for example on the network 22, for download.
• the method allows the seller 4 to follow in detail the delivery of the package 9, since the assumption of responsibility for this package by the carrier 6 until its final delivery to the buyer 11.
• the method also makes it possible to deal with any anomalies of transport such as delay, absence of delivery at the end of a certain delay, package spoliated or deteriorated.
• Another type of anomaly resides in the return of the parcel to the sender 4 on the ground that the addressee does not live at the indicated address (NPAI), since the carrier 6 did not find the addressee. It may be that the delivery address is incorrect or that the carrier has made a mistake. There is an anomaly when this return was not justified. This is to increase the satisfaction of sellers 4 and especially that of buyers 11 while increasing the productivity of the routing.
• the server 10 obtains from the customer 4 via the network 22 data of all its shipments entrusted to the carrier.
• This data recovery can be done in different ways.
• the seller 4 can transmit online to the server 10 an electronic file 21 containing the data. It can take place by filing the file 21 on a website of the controller 2 for example using a suitable mailing form.
• the sending of the data can be done using software specifically developed to cooperate with a software carrier 6. It can be a data repository through a file transfer protocol or FTP (thus denotes a communication protocol for the computer exchange of files on a TCP / IR network).
• FTP file transfer protocol
• it is advantageous that the file 21 has a header line and separators of a predetermined type and integrates all the fields available in the carrier software 6.
• the server 10 checks the data thus obtained before importing them into the database 12.
• criteria to which the transmitted data must respond before being integrated. .
• These are, for example, format criteria such as a number of alphanumeric characters that the data must include. If one of these criteria is not fulfilled, for example if the data comprises a number of alphanumeric characters different from that stated in the criterion, the server 10 signals that the data has such an anomaly and isolates it by mentioning it. in a list. This list is consulted by a person who modifies or corrects the data concerned to remove the anomaly. In this case, the data associated with each package 9 must meet the following criteria:
• the parcel number comprises a predetermined number of digits, for example 13;
• the delivery address has a postal code comprising a predetermined number of digits, for example 6;
• the date of taking charge of the package by the carrier 6 from the seller 4 has a predetermined format (for example of the type "dd // mm / yyyy");
• the vendor data considered valid, possibly after correction, are recorded by the system 8 in the database 12 of the controller 2.
• the server 10 obtains from the carrier 6 data relating to the transported packages 9. It retrieves, for example for each package, the routing data, in particular the routing states, available on the information system 18 of the carrier. Specifically, in this case, during a step 30, the server 10 queries the system 18 to obtain the data in question and, during a step 32, imports these data.
• the server 10 queries the system 18 to obtain the data in question and, during a step 32, imports these data.
• carrier data are also recorded in the database 12.
• the database thus has data on the packages 9 emanating from the seller 4 and data on the same packages from the carrier 6.
• the system 8 then proceeds to an analysis of the data.
• the steps for obtaining the data that have just been presented are carried out periodically and for example every day, or even several times a day.
• the analysis that we will present is carried out. The latter also takes place periodically and preferably every day or even several times a day.
• the server 10 determines during a step 34 if the package has been delivered.
• the delivery in question here is the first presentation of the parcel by the carrier to the buyer's home. If the latter receives the package, it is the final delivery, that is to say the actual handling of the package by the buyer. Otherwise it is an intermediate delivery, the carrier comes to the home of the buyer without being able to reach the latter.
• the server 10 determines in a step 36 if the delivery has occurred after the deadline. For this purpose, it performs in this case a calculation of a period R in days with the following method:
• R delivery date - date taken - number of days not worked on the period concerned.
• the difference between the delivery date and the pick-up date makes it possible to obtain a duration to which the number of days not worked is subtracted.
• the server 10 determines whether the value R is greater than a predetermined threshold, for example a function of a parameter of the package concerned, such as the type of package. If so, the delivery is considered to have taken place after the deadline, which will trigger the preparation and the sending of a claim as will be seen below. Otherwise, this triggering does not take place.
• a predetermined threshold for example a function of a parameter of the package concerned, such as the type of package. If so, the delivery is considered to have taken place after the deadline, which will trigger the preparation and the sending of a claim as will be seen below. Otherwise, this triggering does not take place.
• pending package therefore designates here routing states in which the package has already been the subject of a first presentation to the recipient.
• step 34 the system determines in which category is the package among the following three:
• the parcel has already been the subject of a first presentation to the addressee. He has not had his final delivery and is waiting somewhere (for example at a place of withdrawal by the buyer). This is the case example when a second presentation is planned;
• the parcel continues its routing. For example, it is being prepared for delivery.
• R date of the day - date of taking over - number of days not worked in the period concerned.
• the number of working days elapsed since the assumption by the carrier is calculated.
• the server 10 determines in the next step 49 an estimate of the number of days from which the package is pending.
• the system calculates a period R as follows:
• R date of the day - date of care - number of days not worked in the relevant period + constant.
• the constant is chosen according to a parameter of the package or the carrier, for example among the following values: 4, 8 or 20. It is chosen for example according to the type of routing service which benefits the package. This amounts to doing the same calculation as the previous one but increasing the result with the constant.
• the server 10 determines that the situation requires investigation and prepares a claim. If the result does not exceed this value, the server does not prepare a claim.
• the threshold taken into account depends on a characteristic of the package, for example a parcel parameter such as its type.
• the server may be scheduled to trigger an investigation, depending on the case, 4 or 16 days after the normal time for the initiation of an investigation. It is assumed in the following that it is necessary to send a complaint.
• a claim message 45 is prepared by the server 10 by means of a template or template 44 that the server chooses from models corresponding to predetermined types of situation ( delay, investigation, spoliation, deterioration, parcels not found, etc.).
• Each of these electronic models includes fields that are automatically filled by the server 10 using the package data available in the database 12.
• the claim includes one or more details of the carrier 6.
• step 46 the server 10 transmits to the seller 4, the client of the service offered by the invention, via the network 22 a message that invites it to validate on the controller's website 8 of the draft claim message relating to different packages, listed in a list and to consult these projects.
• the client 4 has two methods. It can proceed to a manual validation for each message, which assumes for the seller 4 to control his server 14 so that it in turn controls the server 10 that sends the message.
• the seller can, for example, validate the sending of each complaint in a dialogue box.
• the seller 4 may otherwise choose a fully automated shipment that allows him to validate the shipment to the carrier 6 of several claims successively and en masse by means of a single initial order. This method avoids a manual confirmation of each claim that is tedious. It is also possible that the sending of complaints is fully automated without even the customer needs to validate this shipment by an order.
• the claim messages 45 are sent by e-mail or by fax, by post, possibly by registered mail, by telephone, etc. It can be expected that the sending takes place so that the carrier 6 receives these messages as emanating from the seller 4 and which possibly do not show the controller, although this shipment was prepared and performed by the server 10,
• a next step 50 it is assumed that the carrier 6 sends a response to the seller 4 which transmits it to the controller 2 or the response is sent directly to the latter.
• the system 8 then performs a computer analysis of the response including the ability to extract data in an automated manner. This will include determining whether carrier 6 has made a positive response to the complaint.
• the documents are scanned in the chain by the means 10, 12 in order to be converted into electronic form by a suitable conventional coding method.
• the rest of the process involves only the electronic version of the documents thus obtained. Digitization takes place for example by means of a high resolution scanner.
• the means 10, 12 For each electronic file corresponding to a document, the means 10, 12 first perform a step of comparing the file with at least two predetermined types of files. When a predetermined condition on the file types is fulfilled, the means perform a step of associating the file of the document with one of these two types. Thus, when the means recognize that the digital file corresponds to a letter-type document, the rest of the method will include steps adapted to this type of file. It is the same when the document is a summary report including a table. Recognition of the file type by the means 10, 12 is of a conventional type and will not be detailed here.
• the means 10, 12 implement a character recognition step, including alphanumeric characters (numbers, letters) or other types of characters in the document.
• the means obtain a reference of at least one article formed of characters and mentioned in the letter. This is in Figure 7 of reference 96 indicated as "12345".
• the package references are standardized so that the automated means can easily identify such a reference in the document.
• these same means also obtain a string of the characters of the content of the letter.
• the means automatically eliminate parts of the document forming the beginning and the end of the letter. This is for example the header of the issuer and the address of the recipient or the signature and what follows.
• Means 10, 12 thus obtain a raw string 98 which states in this case as follows: "Madam, Sir, Following your request of 4 September 2009, here is the package for which we have launched an investigation: ABCD789 Éliane Durant Our reference for this survey is 12345. Please remember this reference in all correspondence. Please accept, Madam, Sir, the expression of our best regards. " This string of characters here includes several sentences.
• the means 10, 12 eliminate from the raw string 98 one or more words or group of words of an irrelevant type.
• the means compare fragments of the chain with words 100 or groups of words 100 of a predetermined type recorded in the base 10, 12 to detect their presence in the chain.
• variables such as names, first names, addresses, dates, etc.
• the means 10, 12 compare the purified chain 102 with models 104 recorded in the database 12. These models are called “documents" in FIG. 8. This comparison is carried out on a number of models that can amount to a few tens, a few hundred or even a few thousand or more. This comparison relates to the chain of characters of the chain 102 with that of the models 104. The comparison of the chain 102 with each of the models 104 gives rise to the calculation of a correspondence rate between them. This rate oscillates between 0% which marks the total absence of correspondence (relatively rare) and 100% which characterizes a perfect or absolute correspondence (relatively rare also).
• the automated means determine whether the refined chain comprises at least one word or a group of words of a predetermined type said prohibited in connection with the model. If one of these words or groups is present, the calculation of the rate does not take place for this model and the means pass to the next model. If none of these words or groups are present in the chain, the rate calculation takes place. Thus, the automated means calculate the rate only when the streamlined chain does not include at least one word or a group of forbidden words in connection with the model. For each model, the word or words and groups of words thus prohibited are defined beforehand.
• forbidden word makes it possible to block the attachment of a character string to a model if a forbidden word is present in the character string (ex: if "bordereau” is present in the string of characters, one does not may not retain a template requesting the commercial invoice and for which the word "slip" is identified as a prohibited word).
• the rate corresponding to each model comprising this word or group of words is increased.
• a word is for example the term "invoice” or "slip".
• each rate concerned is increased by 3%.
• a test step 106 is then performed. During this step, it is first determined whether one of the rates exceeds a predetermined threshold value and identified with a minimum. This value is for example set at 70 or 80%. If none of the rates reaches this value, it is considered that the chain 102 is not like any of the models registered in the database and human intervention is planned. Otherwise, the highest rate is determined among those that exceed the threshold.
• the device operates autonomously that is to say it is able to analyze the document to recognize a situation similar to a known situation and to draw the necessary consequences.
• the process exceptionally requires human intervention to determine whether the processed document can be compared to a situation already known or conversely if it is an entirely new situation that requires the creation of a new model.
• any new case encountered is thus analyzed by at least one person, preferably at least two, in order to determine the actions to be undertaken. It can thus be decided that the channel 102 must be used to create a new model 104 registered in the database. The autonomy of the device is then increased in the management of the responses and actions to be carried out. Admittedly, the human intervention for the creation of a new model may seem restrictive and likely to slow down the start-up or operation of the system according to the number of new cases encountered. However, each model usually refers to tens, hundreds, or thousands of similar cases. These are all cases whose processing can be automated later without human intervention when the processed documents will be closer to the new registered model.
• the means 10 scan the table 92 by identifying as such each of the cells and their contents. It thus locates the package numbers 96 and the PCH date 114 associated therewith and for each line associated with a package identifies a string of characters that will be processed.
• the chain 98 is, in this case, the reason associated with the package. In the examples illustrated in FIG. 7, the chains are therefore for the respective packages the following: "out of time” and "delivered package”.
• Each of these strings is treated as the strings 98 and 102 have been processed in the case of the textual document. It is thus possible to eliminate irrelevant words from the chain in order to purify it.
• the string is compared with predetermined models of the database which furthermore do not include forbidden words or groups of words. These models 104 are this time called "statuses" in FIG. 8.
• the minimum rate is, for example, this time of 90% since relatively short chains are present. One can increase the rate associated with each model if a favorite word is encountered in the chain.
• the models 104 are in the species organized as follows in base 12. Some models 104 are called main models or fathers 104p while others are called secondary models or sons 104s. Each secondary model is attached to a single main model. On the other hand, a main model may have one or more secondary models attached to it as there may be none. Each secondary model is qualified as such insofar as it comprises only part of the characters of the corresponding main model string. Moreover, this main model is the one that comes closest to the main models of the base.
• the rate identified at the end of the calculations is the correspondence rate between the chain 102 and a main model 104p, the latter is associated with the chain and the means 10, 12 perform an action concerning the reference of articles 96 by means of of this model.
• the rate is a match rate between the chain 102 and a secondary model 104s, it is the main model 104p to which the model 104s is attached that is associated with the article reference.
• the means then perform an action concerning this reference by means of the main model thus identified.
• the comparison of the purified string with the templates of the database is performed as follows.
• the system first compares the chain with the main models in order to save resources. If this step makes it possible to select a model among the main models, no comparison with the secondary models takes place. Only if this step does not select a model does the system compare the chain with the secondary models in a second step.
• strings from arrays are relatively short, so there is no risk of degeneration given the high level of correspondence required (90% vs. 70 or 80% for textual records) .
• the set of base models to which the chain 102 is compared in the case of a text document 90 will preferably be different from that used for the comparison in the context of a document 92 in the form of a table. It is the preliminary determination of the type of the document which makes it possible in particular to determine the models of the base which must be used for this comparison. It may be the same for sets of words and groups of forbidden words, preferred and irrelevant. Preferably, these sets are determined by an individual prior to the implementation of the method by the automated means.
• the means obtain at least one article reference formed by characters and at least one character string associated with this reference and comprising the routing status.
• the base 12 includes routing state models in registered form and which form on the one hand principal models or father statuses and secondary models or child statuses which are each attached to a single main model.
• the means calculates a correspondence rate between the chain and each of the routing state models.
• the means determines if at least one of the rates exceeds a predetermined minimum threshold, and if so, if one of the rates is higher than the others. If the highest rate corresponds to a master model, the means associate the article reference to that model and perform an action using that model. If the highest rate is for a secondary model, the means associates with the article reference the main model to which this secondary model is attached and act accordingly. If no rate condition is met so that no pattern can be associated with the routing status, a person is called in to review the situation. This can result in registration in the database of a new routing state template.
• the main routing states are furthermore grouped into respective respective categories:
• - undelivered parcels these are the routing reports which could not be the subject of a first presentation to the addressee.
• - parcels delivered these are the routing states that have already been the subject of a first presentation to the recipient of the package. For routing states considered delivered, an additional classification is performed. This distinguishes the final delivery for the delivery states that confirm the delivery of the goods to the consignee, the "intermediate delivery + 4" status for the routing states that will be the subject of an investigation if the state routing shall not change after a period of 4 days in addition to the investigation period which depends on the type of product, or the "intermediate delivery + 16", with regard to which will be investigated if the routing status does not change after 16 days in addition to the inquiry time which is based on the type of product.
• the data of the carrier obtained from the response is checked by the server 10 and, if anomalies are detected, they are subject to an automatic modification or correction by the server, or else a treatment by a person for example using an error management module that can correct read errors.
• the system 8 analyzes the response and confronts it with the data already present in the base 12.
• a step 60 he prepares a response or challenge 45 to the carrier 6 which he then makes available for validation by the seller 4 in a step 62. As before, once the validated response, the system sends it to the carrier 6 during a step 64.
• all or part of the data associated with a parcel 9 are made accessible remotely and online to the seller 4 by means of a computer document such as a web page.
• the seller can thus view this page from his computer 14 by displaying it on the screen of the latter. It can also be expected that the buyer 11 can also access a similar page 71 from his computer 67.
• the page 66 presents both data 63 relating to the package and from the seller 4 and data 61 relating to the package and from the carrier 6. This is for example the following data or information:
• This page 66 is a form allowing the seller to have quick and centralized access to all the information concerning a package. It can be foreseen that it also gives all possible information in case of delay or loss such as the number of days late, the inquiry number, etc.
• This page also allows the sender 4 to access all the archived documents relating to a package and whose content is included in the summary sheet. He also has access to information enabling him to inform the buyer as soon as possible if necessary.
• this page can be used by the seller to validate the sending of a complaint. But preferably, this page only allows to initiate certain claims of the type deterioration, spoliation, dispute of acquired and NPAI and it is another page which allows the sending of complaints to the unit, en masse or automatically .
• the file thus put online can be made accessible to the services of the carrier 6 or sent to them. (For parcels not found, however, it is up to the seller to verify that the package exists and has been sent before requesting the release of a claim for it.It must confirm or not the shipment of the package.)
• the complaint is then prepared by the server 10 according to models adapted to the cases considered as mentioned above.
• the same interface makes it possible to request the carrier to provide more detailed information on the delivery of the package if necessary.
• the vendor 4 is provided with a direct access URL web address which enables him to directly open the desired parcel form without having to enter an identifier or to search for the parcel data on the controller's website. For this, the system sends the seller an email containing the URL in question.
• the system does not send an email for each package to sellers, but their address a URL that the seller can take into account by adding a package number at the end so that it points directly in the parcel card who interests him from his own facility.
• the system 8 ensures the production and putting on line of the page 66, then its modification for its regular maintenance, for example daily, according to the update of the data relating to the parcels.
• the system 10 makes available online page 71, similar to page 66, for remote consultation by the purchaser 11.
• the latter has extensive follow-up information. It is possible to customize this page for each of the sellers 4.
• the buyer 11 is provided with a direct access URL web address that allows him to directly open the desired parcel form without having to enter an identifier or to search the package data on the website of the controller 2.
• the system sends the buyer an email containing the URL address in question.
• the information on page 71 can be broken down as follows into four groups or panes on the page:
• - Package information 9 such as: the carrier's parcel-related data, the contractual delivery times, the tracking number (for example with a link to the carrier's parcel page), the date of shipment shipping, the consignee and his address;
• the component indicates the expected date of delivery, that is to say the day the package must theoretically be delivered if the carrier complies with the contractual deadlines. This date is purely indicative, the package can be delivered after or even before this time.
• the page indicates to the buyer 11 that an anomaly has been detected during the delivery and that an investigation was initiated by the shipper (using System 8) with the carrier.
• the buyer is invited to contact the customer service of the seller only from a date indicated on the shutter. It is from this date that the sender will be best able to inform the buyer about the anomalies following the shipment; and
• this page 71 allows the buyer to warn the seller of an anomaly. The latter can then request from his own page 66 the preparation and sending of a complaint.
• Vendor page 66 may have similar content.
• the main models are classified in a hierarchical order and identified by a serial number that allows either the creation of a standard response using a template or "template” that is customized according to the needs and proposed to the customers of the controller, or to create a special action to inform these customers about the situation encountered.
• the hierarchical order assigned also makes it possible to follow the progress of the claims which are open with the carrier and which can for example evolve according to the following six states: (0) "no information", (1) “in progress”, (2) “pending document”, (3) “disagreement”, (4) "pending reimbursement”, (5) "definitive refusal” and (6) "refunded”.
• the hierarchical order makes it possible to evolve the state of the complaint towards one of these states and according to the correspondence which is associated with it. For example, models whose hierarchical order number is between 1 and 100 are associated with state (1) "in progress". It is expected that claim statements can never regress so that the outcome of the claim is enforced.
• the means seek a match for the customer number, the slip number and the date of shipment on the slip with the data of the database. If these criteria are met, the bill is integrated into the customer's account. If an error is generated, human intervention is expected.
• each shipment is identified by its tracking number, calculated weight during shipment and the resulting invoice price. (Indeed, when the sender delivers his parcels to the carrier, he declares a weight and a price calculated by his own services, which are then recorded in the system database 8).
• the carrier 6 may address a reimbursement to the seller 4. All necessary data is sent for this purpose by means of the carrier's service process. Once the carrier's response has been received, it is analyzed by the system 8. If the compensation has been granted, the procedure is closed.
• the server 10 crosses the available information again. If it appears that there has been delivery outside contractual deadlines, the server 10 prepares a reminder message to the carrier, offers it to the seller beforehand for validation, then sends it once validated, possibly accompanied by documents justifying this delay.
• the method also makes it possible to analyze the content of these invoices and to highlight for the sender 4 the list of parcels which have been overcharged beyond a configurable threshold.
• This principle is implemented as follows. With reference to FIG. 6, during a step 70, the invoice of the conveyor obtained by the controller 2 is analyzed in order to extract the relevant data in the same way as indicated above, which data are optionally checked or even corrected.
• the data is stored in the database 12 of the system 8.
• the system analyzes whether the billing is excessive compared to what was expected, namely whether the delivery of the package was overcharged or not. For this, it performs the comparison of the seller's data and carrier data.
• step 74 if it appears that the transport has not been overcharged, the analysis ends.
• step 75 the system 8 determines in step 75 whether this question itself requires the preparation of a claim message or not. This determination may in particular obey a choice of the seller 4.
• the system 8 prepares as previously a claim message 45 during a step 76, makes it available to the seller for validation at step 78, then if necessary the address to the carrier 6 to the Step 80.
• the seller can therefore examine whether the extra billing has been justified or not. This confirmation does not occur when opening a simple survey for example, since we have information on the routing status confirming the reality of the shipment.
• each claim file is not closed until the reduction or refund has been ascertained through data relating to an invoice sent by the carrier, unless there are legitimate reasons led to closing the claim.
• the method enables the seller 4 to check the conformity of the carrier's invoice with the data recorded by the system 8.
• the seller 4 can be provided on the site of the controller 2, for example, an interface allowing him to create lists of packages in his opinion overcharged by the carrier. He can in this way if he wishes to request the opening of a claim with the carrier to obtain more information on each of these billing anomalies.
• the process also allows the seller, if he wishes, to engage in turn on delivery times with its buyers, by offering a compensation service through the compensation provided upstream by the carrier since it is he who agrees on delivery times and provides compensation costs shipments in the event of a delay.
• the method according to the invention therefore allows the seller to pass on to the buyer compensation for delivery out of time, if the carrier has not respected its contractual commitments. This repercussion takes place automatically in whole or in part as follows:
• the seller daily loads the list of his shipments on the system 8;
• the system 8 sends by e-mail a shipping notice to the buyer 11 containing the secure URL address that leads to the web page 71, for example a seller's page or dedicated to the latter;
• the buyer 11 can track the delivery history of his package with the expected delivery date, delivery anomalies detected by the system 8, etc.
• the system 8 prepares a claim for compensation from his services as described above;
• the seller sends the buyer a refund notice and proceeds to the reimbursement himself by bank transfer, issue of a check, etc .;
• the carrier carries out the compensation of the parcels out of time with the seller in the form of assets for example.
• the method according to the invention makes it possible to improve the handling of delivery incidents, which treatment is carried out very shortly after the appearance of the incidents without waiting for a possible alert from the buyer concerned.
• the system checks the routing of all parcels shipped, directly using data available on the carrier's website or on its information system without going through the website. of the last.
• the implementation of the method is automatic and does not require the intervention of the seller 4 if it is the transmission of data relating to parcels concerning him and possibly the validation of the sending of the complaints, when they are not performed automatically.
• the sender 4 does not need to wait to be alerted by the buyer in the event of an anomaly. He thus gains precious time. He is also able to inform a buyer of an anomaly that an investigation has already been opened or a complaint sent. The latter feels reassured because he learns that the seller 4 closely controls his shipments and has already intervened upstream to try to solve the problem.
• the method also makes it possible to follow the compensation procedure following the delay or the loss of the package.
• the invention makes it possible to present, in an organized manner, tasks to be performed while making the necessary documents accessible in a dematerialized manner.
• two solutions are possible. If the compensation has been granted, the procedure is put on hold pending repayment.
• the automated means cross again all the information in the database. If it appears for example that there has been delivery outside contractual deadlines, the carrier's response is contested and a reminder is sent to him accompanied by documents justifying the delay. The procedures are thus followed until the closing of the file with a centralization of procedures and parts in the database. The procedures are only closed once the compensation is found on the carrier's invoice.
• the invention enables the constitution in the database of a real memory of the different files by recording the information of the documents successively processed.
• Processing a document generates three actions. The first is to log the scanned document into the database by assigning it with the key information recognized during scanning (document direction, file number, package name, name, etc.).
• the means having interpreted the response of the interlocutor look in the database for the previous documents relating to the current file. This search retrieves the information needed to develop the response.
• the system generates the appropriate response based on the information available. And what the system has assimilated during the procedure can be used later for a similar procedure.

Landscapes

• Business, Economics & Management (AREA)
• Engineering & Computer Science (AREA)
• Economics (AREA)
• Strategic Management (AREA)
• Entrepreneurship & Innovation (AREA)
• Human Resources & Organizations (AREA)
• Quality & Reliability (AREA)
• Operations Research (AREA)
• Marketing (AREA)
• Development Economics (AREA)
• Tourism & Hospitality (AREA)
• Physics & Mathematics (AREA)
• General Business, Economics & Management (AREA)
• General Physics & Mathematics (AREA)
• Theoretical Computer Science (AREA)
• Educational Administration (AREA)
• Game Theory and Decision Science (AREA)
• Management, Administration, Business Operations System, And Electronic Commerce (AREA)
• Devices For Checking Fares Or Tickets At Control Points (AREA)

Applications Claiming Priority (2)

Publications (1)

Family

ID=42113428

Family Applications (1)

Country Status (6)

Families Citing this family (15)

Family Cites Families (20)

• 2009
  • 2009-10-28 FR FR0957582A patent/FR2951846A1/fr not_active Withdrawn
• 2010
  • 2010-10-28 BR BR112012010106-4A patent/BR112012010106B1/pt active IP Right Grant
  • 2010-10-28 WO PCT/FR2010/052319 patent/WO2011051630A1/fr active Application Filing
  • 2010-10-28 EP EP10790653A patent/EP2494470A1/de not_active Ceased
  • 2010-10-28 JP JP2012535907A patent/JP5718347B2/ja active Active
  • 2010-10-28 US US13/504,147 patent/US9330371B2/en active Active

Non-Patent Citations (2)

Also Published As

Similar Documents

Legal Events