Computer-implemented intervention prognosis in production facilities

Abstract

Disclosed is a computer-implemented method for intervention prognosis in a production facility, as well as an associated data processing device and a computer program.

Description

[0001] P101 00112A - 1 -

[0002] Computer-Implemented Interventional Prognosis in Production Plants

[0003] TECHNICAL AREA

[0004] The present invention relates generally to the field of production, preferably chemical production, and in particular to a method for predicting interventions in a production plant, preferably a chemical production plant.

[0005] BACKGROUND

[0006] In the manufacturing industry (e.g., the chemical industry), numerous products are manufactured using batch processes. This typically involves combining several pieces of technical equipment into a single process, with a multitude of processes potentially taking place within a single production plant. Often, each process involves the production of a single product in numerous batches over several months or even years. A batch progresses through this process in a defined, predetermined sequence. The added value of the entire process increases when more batches are produced per unit of time under otherwise identical conditions (batch size, yield, quality).Since the process equipment is planned for each product on a long-term basis, and both the sequence in which a batch passes through the process equipment and the sequence of process steps for each piece of equipment are fixed, an increase in the added value of the production plant cannot usually be achieved by better planning of these sequences.

[0007] The added value and efficiency of a production operation can be increased by optimizing process cycle times. Optimizing process cycle times can be achieved by reducing the average cycle time and by reducing cycle time fluctuations. The latter also leads to improved process predictability.

[0008] One way to achieve this lies in the efficient use of resources. Due to the complexity of such production facilities, in which several processes are interdependent (e.g., regarding the resources used or required for interventions), it is difficult to optimally utilize the available resources. To generate the highest possible added value in the event of resource bottlenecks, current prioritization of processes and equipment is therefore necessary, as well as a high degree of transparency regarding this prioritization, so that resources can be used as efficiently as possible for interventions. There is a need for improved intervention forecasting techniques for use in resource planning.

[0009] US 2019 / 0324432 A1 discloses a complex sensor system in an industrial environment. A data analysis circuit analyzes the sensor values ​​to determine states such as process status or component status. Based on these determined states, actions can be triggered, such as adjusting process parameters, modifying sensor settings, or changing device settings. The goal is to monitor and predict the quality, efficiency, and safety of the plant, for example, through fault prediction. Intervention prediction, particularly of planned interventions for resource planning, is not addressed, nor is automated bottleneck analysis for prioritization and thus efficiency improvement.

[0010] Therefore, an objective of the present invention is to provide a method for predicting interventions in a production plant, preferably a chemical production plant, which at least partially overcomes the aforementioned disadvantages of the prior art.

[0011] SUMMARY OF THE INVENTION

[0012] This is achieved by the subject matter defined in the independent claims. Advantageous modifications of embodiments of the present disclosure are defined in the dependent claims, as well as in the description and the figures.

[0013] One aspect of the invention relates to a method for predicting interventions in a production plant, preferably a chemical production plant. The method can be computer-implemented.

[0014] In the chemical, pharmaceutical, food, cosmetics, semiconductor, biotechnology, plastics, metal, glass, textile, automotive, electrical, and aerospace industries, a "production plant" can be understood as a technical facility used to carry out biological, chemical, physical, and / or process engineering operations for the purpose of manufacturing products typical of these technical fields. P101 00112A - 3 -

[0015] A "chemical production plant" can be understood as a technical facility used to carry out chemical, physicochemical, biochemical, and / or process engineering operations with the aim of producing chemical products, intermediates, formulations, or similar products. Such a plant typically comprises one or more "units of equipment." These units may include specific equipment such as reactors, mixers, destination or rectification columns, evaporators, filters, dryers, and / or tanks.

[0016] A process carried out by a production plant can be a batch process. The term "batch process" (or "batch process") can be understood, for example in the context of a chemical production plant, as a production method in which a finite, defined quantity of raw materials or mixtures of raw materials undergoes a defined sequence of process steps or recipe steps within a limited timeframe to produce an equally finite quantity of intermediate product or product (the batch). Batch processes are typically contrasted with continuous processes, in which, as in the example of the chemical production plant, the substance or mixture flows continuously through the plant, and the process steps occur simultaneously in different parts of the plant, spatially separated and overlapping in time.Batch processes are frequently used in the specialty chemicals, pharmaceuticals, fine chemicals, biotechnology, and food industries, particularly in the production of high-value products, in smaller production volumes, when high flexibility in product variety is required on the same equipment, or when traceability of each individual production batch is important. They are often recipe-driven.

[0017] The term "recipe" can be understood as a step-by-step instruction or manufacturing procedure for carrying out a process in a device. The recipe can define any number of recipe steps and a specific sequence of process steps, phases, actions, parameter values, transition conditions, and / or logical branches.

[0018] The term "recipe step" can be understood as a subset of a recipe. A recipe step can be a process step, the sum of several process steps, or a subset of a process step. Recipe steps can be completed when defined transition conditions are met or remain active if these transition conditions are not met. Recipe steps can, for example, be P101 00112A - 4 -

[0019] Filling the apparatus with 1,000 liters of water,” “Waiting for transfer from apparatus B” or heating the contents of the apparatus to 80°C within 1 hour.”

[0020] The term "transition condition" can be understood as a set of predefined parameter values, such as sensor data, and conditions whose fulfillment triggers the completion of one recipe step and the transition to the next defined recipe step. Transition conditions and the resulting action could be, for example, "If the temperature is below 20 °C, the total fill level is above 80%, and the downstream device is in the recipe step 'Waiting for Dosing', switch to the recipe step 'Transfer to Downstream Device'" or "If the temperature is below 20 °C, the total fill level is above 80%, and the downstream device is not in the recipe step 'Waiting for Dosing', switch to the recipe step 'Waiting for Downstream Device'."

[0021] Similarly, the terms "batch process", "recipe", "recipe step" and "transition condition" can also be understood in relation to production facilities in the pharmaceutical industry, the food industry, the cosmetics industry, the semiconductor industry, the biotechnology industry, the plastics industry, the metal industry, the glass industry, the textile industry, the automotive industry, the electrical industry or the aerospace industry, with the process steps and equipment typical in these areas.

[0022] The process can include receiving production data from the production plant. This production data can include time information about scheduled interventions in sequences for multiple batches passing through a set of equipment in the production plant.

[0023] Production processes, for example in chemical production plants, were previously carried out largely manually. For instance, a production worker might fill a device with a solvent by manually opening the valve of the solvent line, starting a pump, and then switching off the pump and closing the valve after filling. Today, production processes are generally automated or partially automated, so that the individual recipe steps are executed automatically, and planned manual interventions by production workers are usually rare. For example, a production worker might need to intervene in the production process to fill a finished product or to take a quality sample. P101 00112A - 5 -

[0024] The term "intervention" can be understood as an interaction between a person or a machine and a batch process in a (chemical) production plant. This interaction can be anticipated, i.e., planned and regular, occurring, for example, with every batch or every xth batch during normal operation of the production plant. An intervention can be a manual activity, such as manually filling a device with a raw material, manually opening a valve, manually taking a sample, manually checking a fill level or phase separation, or manually emptying a dryer. An intervention can be crucial for the continuation of a device's process, such that the process cannot proceed to the next recipe step without this intervention. Therefore, an intervention can be time-critical for a device and the process carried out within it.

[0025] The term "regular operation" can be understood as the intended, i.e., planned and normal operation of a plant.

[0026] Production data can be received in the form of a stream and / or a log. In the case of a stream, the production data can be recorded or measured (e.g., using appropriate sensors) and sent and received via a stream (e.g., in real time). Thus, the production data can also be understood as measurement data representing the current, physical state of the production plant (i.e., a real, physical production plant in which a real, physical production process is taking place). In the case of a log, the production data can be recorded or measured (e.g., using appropriate sensors), written to a log file (e.g., stored in a database), and retrieved or received from it. The production data can be in a format that includes time information (e.g., a timestamp) associated with each completed (i.e., already performed) intervention.The intervention can be represented by a name (e.g., a description of the activity) and / or an identifier (e.g., a unique ID, an index, a number, etc.). Additionally, the form of the production data can include or indicate the associated sequence (i.e., the sequence into which an intervention has occurred) (e.g., by means of a unique ID, an index, a number, etc.). A sequence as used here can comprise a recipe step, part of a recipe step, or multiple recipe steps, or a repeating pattern of one or more recipe steps (e.g., identifiable by corresponding information of the recipe steps, such as name, index, or a combination thereof). P101 00112A - 6 -.

[0027] The procedure can include defining a prioritization order that specifies the order of priority for the set of equipment and an associated set of processes within the production plant. The set of equipment can include equipment from a single associated process or equipment from different processes.

[0028] A "sentence" as used here can describe a set of one or more elements. In other words, a "sentence" can also be understood as "one or more" or as "at least one".

[0029] The prioritization order indicates the relative importance of the devices. Each device can be assigned a priority. A device with the highest priority can be at the top of the priority order, and a device with the lowest priority can be at the bottom. In other words, the devices can be arranged in descending order according to their priority.

[0030] The priority can depend on the batch time. For example, a bottleneck device (i.e., a device with a long batch time) can have a higher priority than a device with a shorter batch time. Thus, the bottleneck device can be positioned before the device with the shorter batch time in the prioritization sequence. Determining the prioritization sequence can be based on an automated bottleneck analysis of time-coupled devices in a production plant, preferably a chemical production plant, as described in more detail below.

[0031] Within a production line, a batch typically goes through various process steps, each of which can be value-adding or non-value-adding. The duration of these process steps can vary from batch to batch. The sum of the value-adding process steps can be referred to as the batch time of a batch. The batch time of a single machine also fluctuates from batch to batch. The characteristics and reasons for these fluctuations are different and can change dynamically over time. This applies to each individual machine in the process. The sum of the non-value-adding process steps can be referred to as the waiting times of a batch. Such waiting times arise, for example, because resources are not available for the necessary interventions in the processes. Process steps can comprise one or more recipe steps.The cycle time can be defined as the duration of a game's stay in a machine. The cycle time of a machine can fluctuate from game to game. Thus, the average cycle time and the P101 00112A - 7 - are determined.

[0032] Cycle time variation of the entire process from the cycle times and cycle time variations of the individual devices, which in turn result from the batch times and waiting times as well as their variations, which in turn are composed of the recipe step durations as well as their variations.

[0033] However, a production plant can also have multiple processes running simultaneously (i.e., the set of processes includes more than one process) (e.g., concurrently, sequentially, partially overlapping, etc.). Each process is associated with a set of equipment (i.e., as part of a process, a batch passes through every piece of equipment in the set belonging to that process). This can lead to a situation where using individual prioritization sequences in isolation (i.e., ordering the equipment of a process by priority) does not maximize value creation. Instead, a comprehensive overview is needed in which the equipment can be prioritized and represented not only within its own process, but also in relation to equipment from other processes, as well as the individual processes themselves.

[0034] This is made possible by the sophisticated "prioritization definition." In other words, the prioritization definition can be understood as a comprehensive overview or representation of the prioritization order for all equipment across all processes (i.e., across all equipment integrated into the production plant). This is necessary because not all equipment in a higher-priority process necessarily has a higher priority than equipment in a lower-priority process. Rather, it can happen that the most important piece of equipment (e.g., with the highest priority) in a lower-priority process is more important (e.g., in terms of reducing cycle time and thus the productivity of the production plant) than an unimportant piece of equipment (e.g., with the lowest priority) in a higher-priority process.

[0035] The procedure may include determining a currently executed sequence in one apparatus of the set of apparatuses.

[0036] The procedure can include determining predictive information for at least one expected intervention associated with the currently executed sequence, based on temporal information about past interventions. This could be a manual intervention, an intervention by a machine or robot, or the like. The predictive information can be determined for a predefined future period. For example, predictive information can be determined for the next 12 hours, thus indicating expected interventions within that timeframe. A static predictive function (P101 00112A - 8 -) can be used to determine the predictive information.

[0037] A fixed model (i.e., one with fixed parameters) or a dynamic model (e.g., one with variable parameters such as other statistical measures like standard deviation, or taking into account the probability of a sequence occurring in the past) can be used. The dynamic model can incorporate machine learning. Furthermore, in the case of a new sequence (i.e., a sequence being executed for the first time), the duration after the first run can be determined. This duration can then be considered when determining the predictive information.

[0038] The process can include displaying a prioritization definition updated based on forecast information. This display can be shown on an electronic display device (e.g., a monitor, tablet, smartphone, etc.) and can occur during the execution of the currently running sequence. The updated prioritization definition can be used for resource planning for interventions in the production plant.

[0039] Thus, the method provides a solution for analyzing even highly complex production facilities and offering a transparent, objective overview of the production facility through prioritization. In other words, the method provides computer-based support for users in resource planning for the production facility (i.e., targeted human-machine interaction in the technical task of production planning). Based on this prioritization, inefficiencies in the process can be prevented, which might otherwise have occurred due to the complexity and lack of transparency of the process. Accordingly, efficient planning measures for the processes and / or the production facility (e.g., regarding resources for interventions) can be determined based on the displayed prioritization.At the same time, measures based on subjective assessments, but which are inefficient, can be avoided.

[0040] The processes can be recipe-driven. Additionally, they can be pure batch processes and / or processes that may include elements of continuous processes. A batch process comprises at least one piece of equipment with batch operation. A batch process can be linear or branched.

[0041] Apart from chemical production, the method is also suitable for predicting interventions in production plants in other domains such as pharmaceuticals, food industry, cosmetics, semiconductors, biotechnology, plastics industry, P101 00112A - 9 -

[0042] Metal industry, glass industry, textile industry, automotive industry, electronics industry or aerospace industry.

[0043] The described method differs significantly from the very general ideas described in US 2019 / 0324432 A1. For example, the "predictive maintenance" described therein is primarily based on predicting failures or maintenance needs of technical components using sensor data. In contrast, the present invention relates to the prediction of human or machine interventions, more precisely, expected interventions, in batch processes of a production plant, preferably a chemical production plant, based on historical time information about interventions that have already taken place. Thus, the focus is not on the technical condition of equipment or its maintenance cycles, but on the predictive planning of time-critical operating procedures in the regular production process.Furthermore, the present invention enables an automated and dynamic bottleneck analysis of time-coupled devices, which is also not the subject of US 2019 / 0324432 A1.

[0044] In production facilities, such as chemical plants, a "pull" strategy is often used in normal operation. In this strategy, the next batch starts immediately after the previous one is completed, and the production process thus determines resource allocation. Short-term and long-term fluctuations and changes in the production process can lead to dynamic changes in the prioritization order of equipment and make interventions such as sampling or the introduction of solids difficult to plan precisely. All of this can result in production workers being deployed too late, too early, or at the wrong point in the production process, which does not maximize added value.To avoid these inefficiencies, there is a need for a method that identifies bottlenecks in real time and derives a dynamic prioritization sequence, determining and specifying when and where production workers should intervene in the production process. The forecast must not only identify critical interventions but also indicate their likely time windows. This allows production workers to prepare interventions in the production process in a targeted manner, avoid delays and waste, and reduce cycle time fluctuations, ultimately increasing resource efficiency.

[0045] A potential analysis has shown that 4 to 10% of the potential for manual intervention in production processes is wasted for the reasons mentioned above. (P101 00112A - 10 -)

[0046] Simulations have shown that the inventive method, through its dynamic bottleneck analysis and the resulting prioritization sequence, is surprisingly able to leverage these potentials with the same resource requirements and thus make a significant contribution to increasing efficiency.

[0047] It may be assumed that the predicted expected intervention is a planned intervention that is foreseen and therefore expected in the normal operation of the production plant, preferably the chemical production plant.

[0048] It may be planned that the predicted expected intervention will take place regularly, for example with every game or with every xth game in the batch operation of the plant.

[0049] It may be foreseen that the predicted expected intervention includes a manual activity, such as manually filling an apparatus with a raw material, manually opening a valve, manually taking a sample, manually checking a fill level or phase separation, or manually emptying a dryer.

[0050] It may be stipulated that the predicted expected intervention is necessary for the continuation of the process of an apparatus, so that without carrying out the predicted expected intervention, the next recipe step will not be initiated.

[0051] It may be the case that the predicted expected intervention is time-critical for the operation of a device, so that it must take place within a narrow time window in order to continue the process without disruption.

[0052] It may be provided that the forecast information for the predicted expected intervention is based on sequence-specific timing information obtained from a history of interventions in previous games.

[0053] It may be intended that the prioritization definition displayed on an electronic display device serves the purpose of resource planning with regard to the timely availability of operating personnel for the predicted expected intervention.

[0054] It may be possible to determine several predicted expected interventions for an ongoing batch, whereby the predictions may include interventions in different pieces of equipment within the plant. P101 00112A - 11 -

[0055] It may be provided that different types of predicted expected interventions are distinguished and each is predicted separately, for example filling, sampling or emptying.

[0056] It may be envisaged that the predicted expected intervention is determined by evaluating historical intervention cycles across a large number of games, using statistical patterns to predict future interventions.

[0057] It may be intended that the predicted expected intervention is also a machine intervention, which is not triggered, for example, by warnings or indications of sensor errors or maintenance needs, but by planned process steps in the batch process.

[0058] In another aspect, the predictive information can include timing information and associated probability information for the at least one expected intervention. Determining predictive information for the at least one expected intervention can involve determining one or more time intervals based on the timing of past interventions relative to the currently executed sequence. For example, it can be determined how large the time intervals were between the currently executed sequence and past interventions within a previous time period. In other words, if it has been determined that sequence "X" is currently being executed, the time intervals in the past between sequence "X" and subsequent interventions are determined.The intervals can be determined, for example, based on the respective start times of the sequence and the interventions performed. The elapsed time period can be specified using time information that includes the production data (e.g., the last 30 days and / or the last 30 batches).

[0059] Additionally or alternatively, determining predictive information for the at least one expected intervention may include determining a statistical distribution of the specified one or more time intervals in order to obtain the probability information for the at least one expected intervention.

[0060] By providing forecast information that includes not only time information (e.g., a time at which an intervention is expected) but also associated probability information (e.g., there is an 80% probability that the expected intervention will occur at the expected time), irregular P101 00112A - 12 -

[0061] Interventions (e.g., refilling shared trays or emptying sump tanks) can be planned more efficiently.

[0062] In another aspect, the statistical distribution can include a first probability time value, in particular a first quartile, a second probability time value, in particular a median, and a third probability time value, in particular a third quartile. A probability time value indicates a time value of the statistical distribution (e.g., the mean, the median, quartiles, percentiles, etc.).

[0063] Determining predictive information for at least one expected intervention may further include determining an earliest start time and / or end time of the expected intervention based on the first probability time value. Determining the earliest start time may involve determining the elapsed duration of the currently executed sequence, reducing the first probability time value by this determined elapsed duration, and adding the reduced first probability time value to the current time. "Times" as used herein may, for example, refer to a specific time.

[0064] Alternatively or additionally, determining predictive information for at least one expected intervention can further include determining a most probable start time and / or end time of the expected intervention based on the second probability time value. Determining the most probable start time can involve determining the elapsed duration of the currently executed sequence, reducing the second probability time value by this determined elapsed duration, and adding the reduced second probability time value to the current time.

[0065] Alternatively or additionally, determining predictive information for at least one expected intervention may also include determining a latest start time and / or end time of the expected intervention based on the third probability time value. Determining the latest start time may involve determining the elapsed duration of the currently executed sequence, reducing the third probability time value by the determined elapsed duration, and adding the reduced third probability time value to the current time. P101 00112A - 13 -

[0066] In a preferred aspect, determining forecast information of the at least one expected intervention includes determining the earliest start time, the most probable start time, the most probable end time, and the latest end time.

[0067] Determining the aforementioned end times of the expected intervention can be done, for example, by adding the expected duration of the intervention to the corresponding start time of the expected intervention.

[0068] In another aspect, the updated prioritization definition can be displayed using one or more information elements. This display of the updated prioritization definition using one or more information elements can be based on a set of properties associated with the forecast information.

[0069] In another aspect, the set of properties can include a duration of the currently executed sequence, a time interval between the currently executed sequence and a start time of the expected intervention, a duration of the expected intervention, a time interval between the currently executed sequence and a start time of a sequence following the expected intervention, an earliest start time and / or end time of the expected intervention, a most probable start time and / or end time of the expected intervention, a latest start time and / or end time of the expected intervention, a probability of the expected intervention occurring, and / or a priority assigned to an apparatus in the prioritization order for the production plant.

[0070] By adaptively or dynamically displaying the updated prioritization definition, previously hidden optimization potentials within the production plant (e.g., regarding resource planning for interventions) can be more easily uncovered.

[0071] In another aspect, a currently executed sequence is determined for a multitude of devices within the set of devices (i.e., at least two devices) of the production plant. This determines a multitude of currently executed sequences and, consequently, forecast information for a multitude of associated expected interventions. The updated prioritization definition can then be based on this forecast information for the multitude of associated expected interventions. P101 00112A - 14 -

[0072] In another aspect, the (updated) prioritization definition can also be based on a predefined prioritization order between the individual processes of the set of processes and / or cycle times of the individual devices of the set of devices.

[0073] In another aspect, the procedure can include receiving a user command to change the prioritization order. The user command can indicate a prioritization of a first process in the set of processes over a second process in the set of processes, and / or a prioritization of a first apparatus in the set of apparatus over a second apparatus in the set of apparatus. Furthermore, the procedure can include determining a new prioritization order for the processes and / or apparatus based on the user command. Displaying the updated prioritization definition can also be based on the new prioritization order.

[0074] In another aspect, the production data of the production plant can include time information about interventions carried out within a predetermined period, in particular 30 days, and / or a predetermined sample, in particular 30 batches, starting from the current time at which the currently executed sequence is being carried out. Additionally or alternatively, the production data can include and / or correspond to measurement data measured by sensors and / or actuators and / or data interfaces of the processes and / or the equipment and / or the production plant.

[0075] In another aspect, the time interval between a first execution iteration of the procedure and a second, subsequent execution iteration of the procedure can be a predefined duration, in particular one minute.

[0076] This ensures that all relevant production data (e.g., relevant regarding interventions) are used for analysis. This allows for the determination of meaningful forecast information, based on which, and using the corresponding prioritization definition, resource planning measures can be determined. Furthermore, the solution incorporates two self-optimization mechanisms through regular updates using a static mathematical model to determine the forecast information:

[0077] On the one hand, the accuracy of the prediction generally increases the fewer sequences remain before an expected procedure. This gradually eliminates the fluctuations of the elapsed sequences from the calculation. The predicted time points for the earliest, most probable, and latest occurrence of the expected procedure (P101 00112A - 15) thus become increasingly closer together.

[0078] On the other hand, experience shows that more precise planning of a production process leads to more stable sequence durations, which is particularly true for interventions in the process, as these are typically the main cause of fluctuations. Smaller fluctuations, in turn, lead to more reliable forecasts, which in turn make the planning of a production process more accurate.

[0079] In one aspect, the invention relates to a data processing device comprising means for carrying out the methods according to any of the mentioned aspects.

[0080] In one aspect, the invention relates to a computer program or a computer-readable medium on which a computer program is stored, wherein the computer program comprises instructions which, when the computer program is executed by a computer, cause the computer to execute the method according to any of the mentioned aspects.

[0081] BRIEF DESCRIPTION OF THE FIGURES

[0082] The invention can be better understood with the help of the following figures:

[0083] Fig. 1: A flowchart of a method for intervention prediction according to an exemplary embodiment of the present invention.

[0084] Fig. 2: A data processing device according to an exemplary

[0085] embodiment of the present invention.

[0086] Fig. 3: An information element for displaying a prioritization definition according to an exemplary embodiment of the present invention.

[0087] Fig. 4: A display of a prioritization definition according to an exemplary example

[0088] embodiment of the present invention.

[0089] DETAILED DESCRIPTION

[0090] The following section describes representative embodiments illustrated in the accompanying drawings. It should be understood that the illustrated embodiments and the following descriptions refer to examples and are not intended to limit the embodiments to a preferred embodiment.

[0091] Fig. 1 shows a flowchart of a method 100 for intervention prediction. The method 100 can be used, for example, for intervention prediction in a production plant, preferably a chemical production plant. The method 100 can be computer-implemented or computer-aided.

[0092] Procedure 100 may include receiving (step 102) production data from a production plant, which includes time information about interventions made in sequences for a plurality of batches passing through a set of apparatus of the production plant.

[0093] Procedure 100 may include determining (step 104) a prioritization definition (as explained, for example, with reference to Fig. 3) which specifies a prioritization order of the set of apparatus and an associated set of processes of the production plant.

[0094] Procedure 100 can include determining (step 106) a currently executed sequence in an apparatus of the set of apparatuses.

[0095] Procedure 100 may include determining (step 108) predictive information of at least one expected intervention associated with the currently executed sequence based on time information about interventions performed.

[0096] Method 100 can display (step 110) a prioritization definition updated based on the forecast information on an electronic display device during the execution of the currently executed sequence for use in resource planning for interventions in the production plant.

[0097] Procedure 100 may also include any of the aspects mentioned above.

[0098] Fig. 2 shows a data processing device 200 according to an exemplary embodiment of the present invention.

[0099] The data processing device may include means for carrying out the method according to the present disclosure (e.g., method 100). The means may be a processor 202 and a memory 204. The processor 202 and the memory 204 may be operatively connected. A computer program 206 may be stored in the memory 204, wherein the computer program 206 comprises instructions which, when the computer program 206 is executed by a computer or the data processing device,

[0100] 200 is executed, causing the computer to execute the procedure according to any of the mentioned aspects (e.g., procedure 100).

[0101] Fig. 3 shows an information element 300 for displaying a prioritization definition according to an exemplary embodiment of the present invention. The information element 300 represents an expected intervention and is visualized by the rectangle shown with a solid line. The way in which the information element 300 is displayed can be based on a set of properties associated with related forecast information. In the example shown, the display of the information element 300 is based, for example, on the earliest start time 302 of the expected intervention, the latest end time 304 of the expected intervention, the most probable start time 306 of the expected intervention, the most probable end time 308 of the expected intervention, and the duration of the expected intervention (i.e., the length of the horizontally hatched rectangle).the distance between the most probable start time 306 and the most probable end time 308). Additionally, the information element 300 can contain an identifier of the expected intervention 310 (e.g. by means of an ID or description of the expected intervention, e.g. ID = expected intervention).

[0102] Fig. 4 shows a display of a prioritization definition 400a-d according to an exemplary embodiment of the present invention. Fig. 4 shows, based on up to six expected interventions, how a prioritization definition 400a-d is updated over time. Each expected intervention is represented by a corresponding information element 300. One or more information elements 300 can occur per apparatus.

[0103] On the y-axis of the respective prioritization definitions 400a-d, the processes (e.g. by means of an ID such as A and B) and apparatus of these processes (e.g. by means of an ID such as 1, 2, 3 and 4) are shown according to descending priority.

[0104] The x-axis shows a timeline pointing into the future, starting from the current time, which is represented by the left side of rectangle 402.

[0105] Prioritization definition 400a shows the status at 0:00, prioritization definition 400b the status at 2:00, prioritization definition 400c the status at 4:00, and prioritization definition 400d the status at 16:00 of another day (e.g., the following day). P101 00112A - 18 -

[0106] It can be seen how the information elements 300 shown tend to become narrower from prioritization definition 400a to prioritization definition 400c (i.e., the time intervals between the earliest start and latest end times become smaller) as they approach the left side of rectangle 402, as the accuracy of the underlying forecast information increases.

[0107] It can be seen how, from prioritization definition 400a to prioritization definition 400c, the information elements 300 of all expected interventions approach the current time (left side of rectangle 402) at their own individual speeds. This speed of approach is determined by the current progress of the process steps of the apparatus and can differ from apparatus to apparatus and over time. For example, the most probable start time of the expected intervention "Phase Separation Control" shifts from approximately 3:00 a.m. (see prioritization definition 400a) to approximately 5:00 a.m. (see prioritization definition 400c). Thus, a delay has occurred in this apparatus during the period under consideration. In contrast, the most probable start time of the expected intervention "Sample Taking" shifts from approximately 9:00 a.m. (see prioritization definition 400a) to approximately 8:30 a.m. (see prioritization definition 400c).In this apparatus, the process steps were completed faster than expected.

[0108] Once an intervention has been completed, it can be displayed again in the prioritization definition (e.g., as soon as the earliest start time of this expected intervention of a new game is again within the predefined, future period such as the next 12 hours).

[0109] In the prioritization definitions 400a-d shown, the position of devices 1-4 corresponds to their priority. If devices 1-4 belong to different processes (e.g., processes A and B, as shown in the example), their displayed position corresponds to their priority within prioritization definition 400a-d, which, however, does not necessarily correspond to their priority within their own processes. If devices 1-4 belong to the same process, their displayed position corresponds to their priority within the prioritization sequence and thus also within prioritization definition 400a-d.

[0110] In the example shown, apparatus 1 can be a bottleneck apparatus of the highest priority process (e.g., process A) at the depicted time (i.e., the most relevant bottleneck apparatus for the production plant). Apparatuses 2-4 can be further bottleneck apparatuses at this time. Whether an apparatus is a bottleneck apparatus or even the bottleneck apparatus can be determined based on its batch time within the predetermined duration (P101 00112A - 19), in particular 30 days, and / or the predetermined sample size (in particular 30 batches), starting from the current time. Typically, a statistical measure such as the median is used for this purpose. Furthermore, it is also possible to consider the trend of the currently executed sequence for this determination.

[0111] At the time of prioritization definition 400c, with limited resources (e.g., only one production employee in the production plant), value creation can be maximized if the first pending, expected intervention "unloading the dryer" is postponed in favor of the next pending, expected intervention "inserting solids" of the current bottle-neck apparatus 1 and only processed afterwards.

[0112] At the time of prioritization definition 400d, however, apparatus 2 may be a bottleneck apparatus of the process with the highest priority (e.g., process A) (i.e., the most relevant bottleneck apparatus for the production plant at that time). At the time of prioritization definition 400d, with limited resources (e.g., only one production worker in the production plant), value creation can be maximized if the first anticipated intervention, "unloading the dryer," of the current bottleneck apparatus 2 is processed now, followed by the anticipated intervention, "feeding solids."

[0113] If employees of a production plant (e.g., with 100 machines) have to make the decisions illustrated here with limited resources, without the invention described herein, incorrect prioritization can easily occur. This is especially true, but not exclusively, in shift work. The described invention can help reduce these incorrect prioritizations and thereby increase the value creation of the production plant.

[0114] The change in the prioritization definition shown above as an example can be made between any bottleneck devices.

[0115] The prioritization definition can also be updated using a user command to change the prioritization order.

[0116] The term “and / or” used here includes all combinations of one or more of the listed aspects and can be abbreviated with “ / ”.

[0117] Although some aspects related to a device have been described, it is clear that these aspects also constitute a description of the corresponding process, P101 00112A - 20 - where a block or device corresponds to a process step or a feature of a process step. Similarly, aspects described in connection with a process step also constitute a description of a corresponding block, element, or feature of a corresponding device.

[0118] Embodiments of the present disclosure can be implemented on a computer system. The computer system may be a local computing device (e.g., a personal computer, laptop, tablet computer, or mobile phone) with one or more processors and one or more memory devices, or a distributed computing system (e.g., a cloud computing system with one or more processors and one or more memory devices distributed across different locations, such as a local client and / or one or more remote server farms and / or data centers). The computer system may comprise any circuit or combination of circuits. In one embodiment, the computer system may comprise one or more processors, which may be of any type. The term "processor" as used herein can refer to any type of computing circuit, e.g.,a microprocessor, a microcontroller, a CISC (Complex Instruction Set Computing) microprocessor, a RISC (Reduced Instruction Set Computing) microprocessor, a VLIW (Very Long Instruction Word) microprocessor, a graphics processor, a digital signal processor (DSP), a multi-core processor, an FPGA (Field Programmable Gate Array), or any other type of processor or processing circuit. Other types of circuitry that may be included in the computer system could be a custom-designed circuit, an application-specific integrated circuit (ASIC), or similar, such as one or more circuits (e.g., a communications circuit) for use in wireless devices like mobile phones, tablet computers, laptop computers, two-way radios, and similar electronic systems.The computer system may include one or more storage devices, which may comprise one or more storage elements suitable for the application, such as main memory in the form of random-access memory (RAM), one or more hard disks, and / or one or more drives that handle removable media such as compact discs (CDs), flash memory cards, digital video discs (DVDs), and the like. The computer system may also include a display device, one or more speakers, and a keyboard and / or a control device, which may include a mouse, trackball, touchscreen, speech recognition device, or any other device that enables a system user to... P101 00112A - 21 -.

[0119] To enter information into the computer system and to receive information from it.

[0120] Some or all of the process steps can be performed by (or using) a hardware device, such as a processor, a microprocessor, a programmable computer, or an electronic circuit. In some embodiments, some or more of the key process steps can be performed by such a device.

[0121] Depending on specific implementation requirements, embodiments of the present disclosure can be implemented in hardware or in software. The implementation can be carried out using a non-transferable storage medium such as a digital storage medium, for example, a floppy disk, DVD, Blu-ray disc, CD, ROM, PROM, EPROM, EEPROM, or FLASH memory, on which electronically readable control signals are stored that interact (or can interact) with a programmable computer system to execute the respective method. Therefore, the digital storage medium can be computer-readable.

[0122] Some embodiments according to the present disclosure include a data carrier with electronically readable control signals that can interact with a programmable computer system to perform one of the methods described herein. In general, embodiments of the present disclosure can be implemented as a computer program product with program code, wherein the program code serves to execute one of the methods when the computer program product is running on a computer. The program code can, for example, be stored on a machine-readable medium.

[0123] Other embodiments include the computer program for carrying out one of the methods described herein, which is stored on a machine-readable medium.

[0124] In other words, an embodiment of the present disclosure is therefore a computer program with program code for carrying out one of the methods described herein when the computer program runs on a computer.

[0125] Another embodiment of the present disclosure is therefore a storage medium (or a data carrier or a computer-readable medium) on which the computer program for carrying out one of the methods described herein is stored when executed by a processor. The data carrier, the digital storage medium, or the recorded medium is typically tangible and / or non-transferable. Another embodiment of the present disclosure is a device as described herein, comprising a processor and the storage medium.

[0126] Another embodiment of the present disclosure is therefore a data stream or a sequence of signals that represents the computer program for carrying out one of the methods described herein. The data stream or sequence of signals can, for example, be configured to be transmitted via a data communication link, e.g., via the Internet.

[0127] Another embodiment comprises a processing means, e.g. a computer or a programmable logic device, configured or adapted to perform one of the methods described herein.

[0128] Another embodiment comprises a computer on which the computer program for carrying out one of the methods described herein is installed.

[0129] Another embodiment according to the present disclosure comprises a device or system configured to transmit a computer program for carrying out one of the methods described herein to a receiver (e.g., electronically or optically). The receiver may be, for example, a computer, a mobile device, a storage device, or the like. The device or system may, for example, include a file server for transmitting the computer program to the receiver.

[0130] In some embodiments, a programmable logic device (e.g., a field-programmable gate array) can be used to perform some or all of the functions of the methods described herein. In some embodiments, a field-programmable gate array can cooperate with a microprocessor to perform one of the methods described herein. In general, the methods are preferably performed by any hardware device.

Claims

P101 00112A - 23 - REQUIREMENTS 1. A computer-implemented method (100) for predicting interventions in a production plant, preferably a chemical production plant, wherein the method comprises: Receiving (102) production data from the production plant, which includes time information about interventions made in sequences for a plurality of batches passing through a set of apparatus of the production plant; Determine (104) a prioritization definition (400) which specifies a prioritization order of the set of apparatus and an associated set of processes of the production plant; Determine (106) a currently executed sequence in an apparatus of the set of apparatuses; Determine (108) predictive information of at least one expected intervention associated with the currently executed sequence based on the time information about interventions performed; and Display (110) of a prioritization definition (400) updated based on the forecast information on an electronic display device during the execution of the currently executed sequence for use in resource planning for interventions in the production plant.

2. The method of claim 1, wherein the predictive information comprises timing information and associated probability information for the at least one expected intervention; and wherein determining predictive information of the at least one expected intervention comprises: Determining one or more time intervals based on the timing information about the interventions performed in the currently executed sequence, in order to obtain the timing information for at least one expected intervention; and Determining a statistical distribution of the specified one or more time intervals in order to obtain the probability information for at least one expected intervention. P101 00112A - 24 - 3. The method of claim 2, wherein the statistical distribution includes a first Probability time value, especially a first quartile, a second Probability time value, especially a median, and a third Probability time value, in particular a third quartile, and; wherein the determination of forecast information of at least one expected The intervention also includes: Determining an earliest start time and / or end time of the expected Intervention based on the first probability time value; Determining a most probable start time and / or end time of the expected intervention based on the second probability time value; and / or Determining a latest start time and / or end time of the expected intervention based on the third probability time value.

4. The method according to claim 3, wherein determining the earliest start time comprises: Determining the elapsed duration of the currently executed sequence; Reducing the first probability time value by the specific duration that has already elapsed; and Adding the first probability time value, reduced by the time already elapsed, to a current point in time.

5. The method according to any one of claims 3-4, comprising determining the most probable start time: Determining the elapsed duration of the currently executed sequence; Reducing the second probability time value by the specific duration that has already elapsed; and Adding the second probability time value, reduced by the duration already elapsed, to a current point in time. P101 00112A - 25 - 6. The method according to any one of claims 3-5, comprising determining the latest start time: Determining the elapsed duration of the currently executed sequence; Reducing the third probability time value by the specific duration that has already elapsed; and Adding the third probability time value, reduced by the duration already elapsed, to a current point in time.

7. The method according to any one of claims 1-6, wherein the display of the updated prioritization definition (400) is performed by means of an information element (300); and wherein the display of the updated prioritization definition (400) by means of the information element (300) is based on a set of properties which is associated with the forecast information.

8. The method according to claim 7, wherein the set of properties comprises: a duration of the currently executed sequence; a time interval between the currently executed sequence and a start time of the expected intervention; a duration of the expected intervention; a time interval between the currently executed sequence and a start time of a sequence following the expected intervention; an earliest start time and / or end time of the expected intervention; a most probable start time and / or end time of the expected intervention; a latest start time and / or end time of the expected intervention; a probability of the intervention occurring; and / or a priority assigned to an apparatus in the prioritization order for the production plant. P101 00112A - 26 - 9. The method according to any one of claims 1-8, wherein for a plurality of apparatuses of the set of apparatuses of the production plant a currently executed sequence is determined, thereby determining a plurality of currently executed sequences and determining forecast information of a plurality of associated expected interventions; and wherein the updated prioritization definition (400) is based on the forecast information of the plurality of associated expected interventions.

10. The method according to any one of claims 1-9, wherein the updated prioritization definition (400) is further based on a predefined prioritization order between the individual processes of the set of processes and / or cycle times of the individual apparatuses of the set of apparatuses.

11. The method of claim 10, further comprising: Receiving a user command to change the prioritization order, where the user command indicates: a prioritization of a first process of the set of processes over a second process of the set of processes; and / or a prioritization of a first apparatus of the set of apparatus over a second apparatus of the set of apparatus; Determining a new prioritization order of processes and / or devices based on the user command; furthermore, displaying the updated prioritization definition (400) is based on the new prioritization order.

12. Method according to any one of claims 1-11, wherein the production data of the production plant comprise time information about interventions carried out within a predetermined duration, in particular 30 days, and / or a predetermined sample, in particular 30 batches, starting from a current time at which the currently executed sequence is being carried out; and / or wherein the production data comprise and / or correspond to measurement data which were measured by sensors and / or actuators and / or data interfaces of the processes and / or the apparatus and / or the production plant. P101 00112A - 27 - 13. The method according to any one of claims 1-12, wherein the time interval between a first execution iteration of the method and a second, subsequent execution iteration of the method is a predefined duration, in particular one minute.

14. A data processing device (200) which includes means for executing the Method (100) according to any one of the preceding claims 1-12.

15. A computer program (206) or a computer-readable medium on which a computer program (206) is stored, wherein the computer program (206) comprises instructions which, when the computer program (206) is executed by a data processing device (200), cause the latter to execute the method (100) according to any one of the preceding claims 1-12.

Images