Method and system for the automatic detection of productivity of company employees

Abstract

A method for the automatic detection of productivity of company employees, wherein the method comprises at least the following operational steps: acquisition of a first data set through of data acquisition means, which comprises behavioural data of the employee over time; calculation, by means of at least one computer, of a pattern of recurring behaviour of said employee over time on the basis of the behavioural data set; acquisition, by means of the data acquisition means, of a second data set, which comprises relational data of the employee; association to the second data set of a corresponding mood of the employee over time; comparison, by means of a computer, of the behavioural data of the employee with the pattern of recurring behaviours in order to detect first anomalies in such behaviour; comparison, by means of the computer, of the employee's relational data with at least a first pre-set threshold value to detect second anomalies in the relational data; interpolation, by means of the computer, of the first anomalies and the second anomalies to determine a temporal correlation between the first anomalies and the second anomalies and thus generate a third dataset of recurring anomalies over time; comparison, by means of the computer, of the third data set of recurring anomalies with at least a second pre-set threshold value to determine the productivity of the company employee.

Description

METHOD AND SYSTEM FOR THE AUTOMATIC DETECTION OF PRODUCTIVITY OF COMPANY EMPLOYEESDESCRIPTIONField of application of the invention

[0001] The invention relates to a method for the automatic detection of productivity of company employees, advantageously usable in any company in any technical field, in accordance with the preamble of independent claim no. 1 .

[0002] Furthermore, the present invention relates to a system for the automatic detection of productivity of company employees, advantageously operating according to the aforementioned method, in accordance with the preamble of independent claim no. 10.

[0003] More specifically, the method in question finds advantageous use in the company's IT organisation, suited to support the management and administration of the company itself, in particular by monitoring the productivity of employees, whether they are employees or consultants.

[0004] Therefore, the present invention finds its place in the technical field of production and marketing of automation systems for the IT management of the company, in particular in the technical field of automatic and / or automated management of company staff and employees.State of the art

[0005] In any area of technology, it has long been a necessity for companies to employ trained and productive specialised employees so that the company can work optimally.

[0006] However, in many specialised sectors, hiring and training employees is timeconsuming and costly for employers.

[0007] Often - especially in recent years - employees who are dissatisfied with a certain job role can decide to leave it, thus forcing the employer to find and train a new employee, giving rise to a burdensome turnover of staff. The replacement of staff is also known in jargon with the English term “turnover”.

[0008] As is known, there are two types of staff turnover: the so-called "physiological" turnover, i.e. , the natural turnover that occurs in any company reality as a result of normal events that occur in the life of employees, such as retirement, early retirement, leaves of various kinds, etc.; and the so-called "pathological" turnover, i.e., the turnover due to reasons that arise within a work context that is not satisfactory for the employee.

[0009] A worker who is dissatisfied with the job role held in the company usually spends a period of low productivity and apathy within the company itself before resigning, defining a loss not only in terms of time but also economic for the employer.

[0010] Dissatisfaction can arise for a variety of reasons, whether they are merely technical or human, such as, for example, relational difficulties with superiors, a job considered monotonous or due to a lack of appreciation of the employee, such as, for example, a salary considered to be too low.

[0011] As mentioned above, a dissatisfied employee is less performing in completing assigned tasks; it results in a lower productivity and an economic loss for the employer.

[0012] Currently, in the technical field of reference, some methods are known to detect and / or evaluate the productivity of an employee, with the aim of detecting the productivity and reliability of the employee and - if necessary - intervening to mitigate any dissatisfaction.

[0013] Normally, the known methods preliminarily provide for a human control by superiors on the work and productivity of the employees under their control, with the aim of detecting any drops in productivity and / or quality of the work performed.

[0014] Subsequently, the known methods provide for verbally clarifying together with the presumed dissatisfied employee what the problems are in order to arrive at a solution and re-establish job satisfaction that allows the employee to increase his productivity.

[0015] Obviously, the known method has been proved to be not without drawbacks in practice.

[0016] The main drawback lies in the difficulty for superiors to accurately and consistently detect the behaviours of employees that could be a symptom of dissatisfaction, leading to low productivity and - in the long term - resignations of the employee even before the employers could notice the discontent of the employee themselves.

[0017] A further drawback lies in the fact that the failure to detect the symptoms of dissatisfaction leads employers to limit any resignations by budgeting a fund intended to cover the costs of research and training of new staff in the event of resignations.

[0018] In fact, companies normally calculate, based on statistical studies of employees’ resignations, a probable average cost that the company will be required to sustain in each fiscal year for the research, hiring and training of new staff, in the view of probable number of resignations by current collaborators.

[0019] Obviously, the expenditure forecast and the contextual budget allocation is not optimal for the company and suffers from all the main and typical drawbacks of statistical studies, i.e. , a not high reliability which translates into an incorrect use of the company budget.Presentation of the invention

[0020] The present invention intends to overcome - all, or at least in part - the drawbacks of the above-mentioned prior art by providing a method for the automaticdetection of the productivity of company employees.

[0021] In particular, the main purpose of the present invention is to provide a method for the automatic detection of the productivity of company employees that is safe and reliable.

[0022] A further purpose of the present invention is to provide a method for the automatic detection of the productivity of company employees that allows the job dissatisfaction of an employee to be quickly identified.

[0023] Another purpose of the present invention is to provide a method for the automatic detection of the productivity of company employees that allows the company to decide how to approach the dissatisfaction of an employee.

[0024] A further purpose of the present invention is to provide a method for the automatic detection of the productivity of company employees that is substantially automatic.

[0025] Another purpose of the present invention is to provide a method for the automatic detection of the productivity of company employees that is substantially free from human errors in the evaluation of behaviour.

[0026] Furthermore, a not least purpose of the present invention is to provide a method for the automatic detection of the productivity of company employees that is quick and easy to implement.

[0027] Furthermore, a not least purpose of the present invention is to provide a method for the automatic detection of the productivity of company employees that is economical and / or that allows for savings compared to known methods.

[0028] These purposes, together with others that will be better clarified below, are achieved by a method for the automatic detection of the productivity of company employees of the type in accordance with claim 1 .

[0029] Other purposes that will be better described below are achieved by a method for the automatic detection of the productivity of company employees in accordance with the dependent claims.

[0030] According to a further aspect of the present invention, a system for evaluating the productivity of company employees of the type in accordance with claim 10 is provided. Detailed description of the invention

[0031] The present invention concerns a method for the automatic detection of the productivity of company employees as well as a system for the evaluation of the productivity of company employees suitable for implementing said method. The method according to the present invention can be advantageously used in any company context, in any field of technology. In the following, advantageous reference will be made to a context in which the employees work by using an electronic terminal, for example a personal computer or similar, for example in an office. However, the method accordingto the present invention must be understood as applicable also in different work contexts, in which the employees use machine tools, for example in manufacturing or production companies.

[0032] The present invention can be implemented in other variants all falling within the scope of the inventive features claimed and described; such technical features can be replaced by different technically equivalent elements and the materials used; the shapes and dimensions of the invention can be any as long as they are compatible with its use.

[0033] The method for the automatic detection of the productivity of company employees, the subject of the present invention, can be conveniently implemented by using a CPU or a computer.

[0034] The term CPU or computer shall be understood, pursuant to the present patent, to mean any machine intended for the electronic processing of data according to procedures defined through programs or algorithms carried out automatically by the machine itself.

[0035] The computer preferably comprises at least one microprocessor or the like, capable of executing the commands contained in the aforementioned program or algorithm to generate output signals or data.

[0036] According to the present invention, the method in question comprises at least the following operational steps.

[0037] Advantageously, the method according to the invention is intended to be implemented in a substantially automatic manner, i.e. , with minimal human interventions during its execution.

[0038] In this way, the method according to the invention allows to detect and evaluate the productivity of the employees in a rapid and simple manner, overcoming the drawbacks of the prior art cited above.

[0039] Preferably, the operational steps that define the method according to the present invention can be performed in the temporal order in which they are described in detail below.

[0040] Otherwise, the operational steps can also be performed in a different order or sequence than those described, without thereby departing from the scope of protection of the present patent.

[0041] Preliminarily, the method according to the invention provides for a step a) of provision of data acquisition means.

[0042] Advantageously, the data acquisition means may be of any type per se known to the person skilled in the art. For example, the data acquisition means may comprise one or more audio and / or video sensors.

[0043] Preferably but not limited to, the data acquisition means may comprise at least aportion of a computer program or algorithm present inside an electronic terminal used by the employee.

[0044] Alternatively, the data acquisition means may provide for acquisition through the provision of information directly by the employee, wherein for example the information comprises working hours and billing data or similar, as will become clearer below.

[0045] More specifically, the data acquisition means may comprise a page (for example on the Internet or intranet) configured to be filled with the work data of each employee.

[0046] Preferably, the work data comprises working hours, time spent on each assigned operation, distinction between activities that bring profit to the company (billable activities) and activities that are a cost to the company (training, internal management, etc.).

[0047] The method also provides for at least one step b) of provision of at least one computer, in data communication with said data acquisition means and configured to receive said data, process them and send corresponding output data.

[0048] As anticipated above, the aforementioned at least one computer comprises at least one processing module, for example a microprocessor or a microcontroller, configured to receive the data acquired by the data acquisition means, to process them and to send corresponding output data. Preferably, the output data are processed data and comprising additional or different information with respect to the data received by the data acquisition means.

[0049] The method also comprises at least one step c) of acquisition of a first data set through said data acquisition means, said first set comprising behavioural data of the employee over time.

[0050] Advantageously, said first data set comprises temporal information on the activities carried out by said employee.

[0051] More specifically, the first data set preferably comprises a set of information on the activity performed by the employee during at least one working day.

[0052] Preferably, the first data set comprises temporal information, in particular defined during at least one working day, of some (preferably all) of the activities and / or operations performed by the employee.

[0053] Preferably, the first data set also comprises qualitative information of the aforementioned activities, in particular to define which activities performed generate income (i.e. billable activities) and which activities instead define therefore an expense (i.e. non-billable activities, where the salary of the employee is an expense for the employer).

[0054] In accordance with the preferred (but not limiting) embodiment of the present invention, the first data set is acquired (in particular by said data acquisition means) in arecurrent and / or recursive manner.

[0055] For example, the first data set is acquired in a recurrent manner on a daily basis and comprises a list with a time scan of activities carried out by the employee during each working day.

[0056] Advantageously, in other words, said acquisition step c) is carried out cyclically to continuously acquire (for example every working day) a new said first data set.

[0057] Conveniently, the method which is the subject of the present invention also comprises at least one calculation step d), by means of said at least one computer, of a pattern of recurring behaviours of said employee over time on the basis of said behavioural data set.

[0058] Advantageously, said pattern of recurring behaviours calculated in said step d) can be updated in a recursive manner with said new first data set.

[0059] Preferably, the pattern of recurring behaviours is calculated by and / or on the basis of the implementation of a probability distribution algorithm.

[0060] In accordance with the preferred embodiment of the present invention, said calculation of said pattern of recurring behaviours in said step d) is carried out by recursively constructing a normal distribution curve of at least one time instant of a working day of said employee.

[0061] Preferably, said calculation of said pattern of recurring behaviours in said step d) is carried out by recursively constructing a normal distribution curve of each time instant of all the working days of said employee.

[0062] In other words, step d) allows to generate a set of data suited to link temporally random variables (i.e. particular behaviours) around a single mean value.

[0063] For example, the pattern of recurring behaviours calculated in said step d) may bring to light a behavioural pattern of the employee reiterated over time in a plurality of working days, for example in all (or almost all) working days.

[0064] The pattern of recurring behaviours calculated in step d) is intended to act as a base (or logical zero, or offset) for the subsequent steps of the method in question.

[0065] In other words, the pattern of recurring behaviours calculated in step d) preferably defines the behavioural model of the employee during his normal working days, on which any future anomalies will be highlighted.

[0066] Preferably, at each new daily iteration of step d), the pattern of recurring behaviours is updated and / or modified on the basis of the new first data set.

[0067] Preferably, step d) provides for calculating the pattern of recurring behaviours according to a probability density function, in particular by distributing the first data set according to a distribution in a gaussian curve.

[0068] For example, the pattern of recurring behaviours may comprise a plurality ofprobability density functions, wherein each function defines the most probable activity that the employee carries out at each instant of time during the working day.

[0069] The method according to the invention also comprises a step e) of acquisition, by means of said data acquisition means, of a second data set, said second set comprising relational data of the employee.

[0070] Advantageously, as will be described in more detail below, the step e) of acquisition provides for analysing and acquiring (in the second data set) each internal corporate communication within the company or with the outside of the company by means of an analysis of the verbal and / or para-verbal language to establish and / or deduce a mood of the employee who communicates.

[0071] The method according to the invention further comprises a step f) of association a corresponding mood of the employee over time with said second data set.

[0072] Advantageously, said association with said second data set of a corresponding mood of the employee is achieved by using at least one first artificial intelligence algorithm.

[0073] Said first algorithm is preferably continuously trained with said second data set in a recursive manner.

[0074] Preferably, said first algorithm also processes said second data set in a recursive manner.

[0075] Furthermore, said first algorithm preferably returns an output data set comprising a correlation of at least one mood of said employee with said second data set.

[0076] In other words, each internal corporate communication or with the outside, is analysed by the first artificial intelligence algorithm (known per se to the person skilled in the sector and therefore not described in further detail in this description) which establishes the tone of the communication, for example whether this communication is serene, aggressive, denoting anger or even happiness, controversy etc., and saves this information in a database anonymizing the second data set with associated different moods of the employee over time.

[0077] Operationally, the acquisition step e) involves for example taking the second data set from the main corporate communication tools (for example e-mail, calendar, instant messaging, video calls and / or images and / or sounds from the webcam).

[0078] Preferably, the step e) associates each data of the second data set comprising at least one communication from the employee with at least one mood of the employee.

[0079] Operationally, the association step f) preferably associates different relational data to the second data set by using the first artificial intelligence algorithm. For example, from the data acquired by the webcam the first artificial intelligence algorithm is configured to extrapolate information regarding the mood of the employee by recognizingdistinctive traits of behaviour, facial expressions of para-verbal language (for example gestures, sighs, etc.).

[0080] The first artificial intelligence algorithm preferably returns a matrix that contains the time indication and the recipient / sender of the communication, wherein for each communication is associated a value of the mood of the employee.

[0081] Preferably, step e) and f) are carried out one after the other in a cyclical and recursive manner, throughout the employee's working day and preferably for each working day of the employee.

[0082] Preferably, the method involves a step of provision of electronic memory means, placed in data communication with the computer and configured to save and / or store the first data set and / or the second data set and / or the second data set with associated values on the mood of the employee.

[0083] For example, the electronic memory means may comprise a hard disk, a solid state disk (SSD) or any other type per se well known to the person skilled in the art and suited to be connected to a computer.

[0084] The method according to the present invention also comprises a step g) of comparison, by means of said computer, of said behavioural data of the employee with said pattern of recurring behaviours to detect first anomalies in said behavioural data.

[0085] More specifically, the step g) involves detecting activities carried out by the employee that do not fall within the probability distribution calculated during the calculation step d).

[0086] For example, in the case that during step g) an activity is prolonged for a long time in a way that does not correspond to the pattern of recurring behaviours, said activity will be recorded during step g) as a first anomaly of the behavioural data.

[0087] For purely illustrative purposes, if the pattern of recurring behaviours defines that the employee checks emails for about half an hour in the morning and this activity suddenly begins to last much longer, this activity will be recorded as a first anomaly of the behavioural data.

[0088] The method also comprises a step h) of comparison, by means of said computer, of said relational data of the employee with at least a first preset threshold value to detect second anomalies in said relational data.

[0089] More specifically, step h) allows to detect which relational data fall within the normal sphere of moods of an employee and which instead are alarming moods that could be risky.

[0090] The threshold value preset in step h) can be foreseen for each mood and can allow to detect in which particular situation the communication is risky. For example, a communication that is too aggressive (i.e. aggressive above a predetermined thresholdvalue and identified by the first artificial intelligence algorithm) will be identified as a second anomaly in said relational data.

[0091] Similarly, an attitude that is too apathetic and / or lazy (i.e. above a predetermined threshold value) during the working day will be identified as a second anomaly.

[0092] The method according to the invention further comprises a step i) of interpolation, by means of said computer, of said first anomalies and of said second anomalies to determine a temporal correlation between said first anomalies and said second anomalies and thus generate a third data set of anomalies recurring over time.

[0093] Advantageously, in accordance with the preferred but non-limiting embodiment of the present invention, said interpolation of said first anomalies and said second anomalies is achieved by using at least one second artificial intelligence algorithm.

[0094] Preferably, said second algorithm is continuously trained in a recursive manner with said first anomalies, said second anomalies and at least one respective time reference.

[0095] Preferably, said second algorithm processes in a recursive manner said first anomalies and said second anomalies.

[0096] Preferably, said second algorithm returns an output data set comprising a temporal and / or causal correlation between said first and said second anomalies at a corresponding given moment in time to thus generate said third data set of anomalies recurring over time.

[0097] More specifically, step i) of interpolation involves connecting first anomalies at a given instant in time to second anomalies at the same instant in time, in particular to allow them to be linked by causality and / or correlation.

[0098] The method also comprises a step j) of comparison, by means of said computer, of said third data set of recurring anomalies with at least a second preset threshold value to determine the work productivity of said company employee.

[0099] In this way, the method according to the invention allows to overcome the drawbacks of the prior art, providing an innovative method that is fast, rapid and safe.

[0100] In fact, the method is substantially automatic, and has the technical advantage of constantly evaluating the productivity of company employees while allowing employers to detect any anomalies in order to intervene, if deemed appropriate.

[0101] More specifically, the second artificial intelligence algorithm analyses the discrepancies and highlights the anomalies to signal any necessary interventions.

[0102] For example, following the comparison step j), different dissatisfaction scenarios can be identified.

[0103] A first dissatisfaction scenario can be defined by second very critical relational anomalies that occur over a long period of time.

[0104] A second dissatisfaction scenario can be defined by second relational anomalies of low criticality (i.e. slightly above the second preset threshold value) that intersect at the same time instants with a low productivity.

[0105] A third dissatisfaction scenario can be defined by second relational anomalies of high criticality that intersect at the same time instants with high productivity.

[0106] A fourth dissatisfaction scenario can be defined by second relational anomalies comprising sudden changes in mood that occur alone and / or in correspondence with changes in behaviour patterns.

[0107] Advantageously, the method of the present invention also comprises an operational step k) of comparison of said third data set of recurring anomalies with a value and / or hourly cost of said employee to determine a value and / or economic loss of said recurring anomalies.

[0108] For example, the comparison step k) provides for comparing in particular the time periods in which the anomalies of the third set of anomalies occur with the billing data and / or the data of the employee cost in the same time periods.

[0109] In this way, the method according to the present invention, in addition to detecting and evaluating the productivity of the employees, can also calculate and evaluate the economic loss of the company due to a non-optimal mood or in any case outside a threshold range.

[0110] For example, starting from an internal company management system (i.e. from the analysis of invoices and cost centers) and from an analysis of the hourly and / or working reporting of individual employees (e.g. a system in which employees enter billable activities) the method according to the invention allows to generate not only the third data set of anomalies to evaluate the productivity of each employee but also the activities carried out and / or not carried out and therefore the possible economic loss deriving from the activities not carried out or carried out with low productivity.

[0111] Step k) preferably also provides for comparing the third data set of anomalies with initial project plans to establish whether the activities in which the employee acted with low productivity and low operational efficiency are activities that can be defined as scheduled, unexpected or out-of-budget activities.

[0112] In this way, the method according to the present invention allows to detect and qualitatively and quantitatively evaluate the productivity of each employee.

[0113] Advantageously, the method according to the invention also comprises an operational step I) of generation and sending of an alarm signal referring to the presence of recurring anomalies in the productivity of said employee over time.

[0114] For example, if the method according to the invention detects an anomalies falling within one of the scenarios reported above for example purpose, the methodprovides that the computer sends an alarm signal to allow a possible human resources manager within the company to intervene, if deemed appropriate.

[0115] The present invention also comprises a system for evaluating the productivity of company employees, suitable for implementing the method described so far.

[0116] All the features described with particular reference to the method for the automatic detection of the productivity of company employees must be understood to be described - taken alone or in any combination thereof - also with reference to the system for the evaluation of the productivity of company employees.

[0117] The system for the evaluation of the productivity of company employees comprises data acquisition means and at least one computer, in data communication with said data acquisition means and configured to receive said data, process them and send corresponding output data.

[0118] In accordance with what is described with reference to the method, advantageously, the data acquisition means may be of any type per se known to the person skilled in the art. For example, the data acquisition means may comprise one or more audio and / or video sensors.

[0119] The system also comprises memory means operationally connected to said at least one computer and comprising within it at least one program for computer.

[0120] According to the invention, the computer program comprises a plurality of instructions stored in said memory means and executable by said computer, said plurality of instructions being suitable for implementing the method for the automatic detection of the productivity of company employees, also the subject of the present invention and described above in detail.

[0121] In accordance with what is described with particular (but not limitative) reference to the method, the electronic memory means may comprise a hard disk, a solid state disk (SSD) or any other type per se well known to the person skilled in the art and suitable for being connected to a computer.

[0122] From the above description it is therefore evident that the method and the system object of the present invention achieve the intended purposes.

[0123] The present invention can be realized in other variants all falling within the scope of the inventive features claimed and described; these technical features can be replaced by different technically equivalent elements and the materials used; the shapes and dimensions of the invention can be any as long as they are compatible with its use.

Claims

CLAIMS1. Method for the automatic detection of productivity of company employees, wherein the method comprises at least the following operational steps: a) provision of data acquisition means; b) provision of at least one computer, in data communication with said data acquisition means and configured to receive said data, process it and send corresponding output data; c) acquisition of a first data set through said data acquisition means, said first data set comprising the employee's behavioural data over time; d) calculation, by means of said at least one computer of a pattern of recurring behaviours of said employee over time on the basis of said behavioural data set; e) acquisition, by means of said data acquisition means, of a second data set, said second set comprising relational data of the employee; f) association to said second data set of a corresponding mood of the employee over time; g) comparison, by means of said computer, of said behavioural data of the employee with said pattern of recurring behaviours to detect first anomalies in said behavioural data; h) comparison, by means of said computer, of said relational data of the employee with at least a first preset threshold value to detect second anomalies in said relational data; i) interpolation, by means of said computer, of said first anomalies and said second anomalies in order to determine a temporal correlation between said first anomalies and said second anomalies and thus generate a third data set of recurring anomalies over time; j) comparison, by means of said computer, of said third data set of recurring anomalies with at least a second preset threshold value to determine the productivity of said company employee.

2. Method according to claim 1 , characterised in that it further comprises an operational step: k) comparison of said third data set of recurring anomalies with a value and / or hourly cost of said employee to determine a value and / or economic loss of said recurring anomalies.

3. Method according to claim 1 or 2, characterised in that said acquisition step c) is cyclically performed to continuously acquire a new said first data set; said pattern of recurring behaviours calculated in step d) is updated in arecursive manner with said new first data set.

4. Method according to one or more of the preceding claims, characterised in that said calculation of said pattern of recurring behaviours in said step d) is performed by recursively constructing a normal distribution curve of at least one time instant of a working day of said employee.

5. Method according to one or more of the preceding claims, characterised in that said calculation of said pattern of recurring behaviours in said step d) is performed by recursively constructing a normal distribution curve of each time instant of all the working days of said employee.

6. Method according to one or more of the preceding claims, characterised in that said first data set comprises temporal information about the activities performed by said employee.

7. Method according to one or more of the preceding claims, characterised in that said association to said second data set of a corresponding mood of the employee is achieved by using at least one first artificial intelligence algorithm, wherein: said first algorithm is continuously instructed in a recursive manner with said second data set, said first algorithm processes said second data set in a recursive manner, and said first algorithm returns an output data set comprising a correlation of at least one mood of said employee with said second data set.

8. Method according to one or more of the preceding claims, characterised in that said interpolation of said first anomalies and said second anomalies is achieved by using at least one second artificial intelligence algorithm, wherein: said second algorithm is continuously instructed in a recursive manner with said first anomalies, said second anomalies and at least one respective time reference, said second algorithm processes said first anomalies and said second anomalies in a recursive manner, and said second algorithm returns an output data set comprising a temporal and / or causal correlation between said first and second anomalies at a corresponding given moment in time to generate thus said third data set of recurring anomalies over time.

9. Method according to one or more of the previous claims, characterised in that it further comprises an operational step:I) generation and sending of an alarm signal referring to the presence of recurring anomalies in the productivity of said employee over time.

10. System for the automatic detection of productivity of company employeescomprising: data acquisition means; at least one computer, in data communication with said data acquisition means and configured to receive said data, process it and send corresponding output data; memory means operatively connected to said at least one computer and comprising at least one computer program therein; said computer program comprising a plurality of instructions stored in said memory means and executable by said computer, said plurality of instructions being suitable for implementing the method for the automatic detection of the productivity of company employees according to one or more of the preceding claims.