{"id":42603,"date":"2026-08-17T17:15:39","date_gmt":"2026-08-17T15:15:39","guid":{"rendered":"https:\/\/www.logicline.de\/?p=42603"},"modified":"2026-08-16T17:45:06","modified_gmt":"2026-08-16T15:45:06","slug":"machine-data-service-order","status":"publish","type":"post","link":"https:\/\/www.logicline.de\/en\/machine-data-service-order","title":{"rendered":"If machine data only triggers alarms but does not generate a service order"},"content":{"rendered":"\t\t<div data-elementor-type=\"wp-post\" data-elementor-id=\"42603\" class=\"elementor elementor-42603 elementor-42595\" data-elementor-post-type=\"post\">\n\t\t\t\t<div class=\"elementor-element elementor-element-7c03afc0 e-flex e-con-boxed e-con e-parent\" data-id=\"7c03afc0\" data-element_type=\"container\" data-e-type=\"container\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-61419d4e elementor-widget elementor-widget-text-editor\" data-id=\"61419d4e\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<strong>At 2:14 a.m., a unit in the field reports a temperature spike.<\/strong> The value is sent to an IoT application, where a red tile appears. No one is on call for this tile because it isn&#8217;t a service process. By morning, the value is back in the green range, the tile disappears, and no one knows whether it was an early warning or measurement noise.\n\nThree weeks later, the same assembly fails\u2014this time while the system is in operation at the customer\u2019s site. The call turns into an emergency; the technician heads out without the right spare part, and in hindsight, someone realizes that the warning signal had been there all along. This is precisely where predictive maintenance initiatives fail in everyday practice: the bottleneck is rarely the sensor and often not the predictive model, but rather the process in between.\n\nFirst, here&#8217;s what matters:\n<ul>\n \t<li><strong>A threshold value by itself does not constitute an incident.<\/strong> Only a condition that persists over a defined period of time distinguishes the actual event from a false alarm.<\/li>\n \t<li><strong>An alarm needs a responsible recipient.<\/strong> As long as it ends up in a separate application, there is no order, no deadline, and no SLA clock.<\/li>\n \t<li><strong>Without the machine context, the alarm is just a number.<\/strong> Service history, configuration, contract status, and spare parts procurement decide whether the service call goes smoothly the first time.<\/li>\n \t<li><strong>Getting started doesn&#8217;t require a sensor-driven push.<\/strong> Operating hours and cycle counters provide reliable triggers before any retrofit begins.<\/li>\n<\/ul>\n<h2>Three Gaps Between the Measured Value and the Service Order<\/h2>\nMany manufacturers have long since implemented data connectivity. The machines send data, a dashboard displays graphs, and yet little has changed in terms of service procedures. The reason lies in three gaps that occur one after another.\n<h3>Relevance: What exactly is an incident?<\/h3>\nA threshold value is quick to define and just as quick to generate false alarms. Start-up procedures, load changes, and ambient temperatures produce spikes that do not indicate any need for service. If each of these becomes a support ticket, the team will learn to ignore alarms within a few weeks, and the most costly consequence sets in: the real signal gets lost in the noise.\n\nThe analysis only becomes reliable when a persistence condition is added. Not \u201ctemperature above 85 degrees,\u201d but \u201ctemperature above 85 degrees, continuously for more than ten minutes, while the system is running.\u201d This single addition decides whether your service team takes the alert seriously.\n<h3>Plot: Who will take over, and by when?<\/h3>\nThe second gap is organizational. An alert in an IoT application has no assigned person, no deadline, and no status. It is a notification, not a task. As a result, it lacks everything a service organization needs to do its job: assignment, scheduling, an escalation path, and a timer tracking the contractually agreed-upon response time.\n<h3>Context: What does the technician know when he opens it?<\/h3>\nThe third gap results in repeat visits. An alarm with a fault code and timestamp provides no information about the system\u2019s history. Whether the same module was replaced eight months ago, which software version is running, whether a service contract with a four-hour response time is in effect, and which spare part fits this configuration\u2014all of this information is stored in other systems. Without this integration, the technician starts out with just a number.\n\nIn practice, this is what decides the first-time resolution rate. The difference between a work order that includes only the fault code and one that includes the last three service calls for that system, the installed software version, and the appropriate spare parts kit is a second trip to the site. For a fleet in the three- or four-digit range, this adds up to a significant amount that becomes evident in service planning.\n<div style=\"border: 1px solid #e00817; background: #fdf5f6; padding: 18px 22px; margin: 28px 0; border-radius: 4px;\">\n<p style=\"margin: 0;\"><strong>Where are your machine signals getting stuck today?<\/strong>\nIn 30 minutes, we\u2019ll walk you through one of your product lines: which signals are already present, which rules turn them into a valid incident, and what\u2019s missing to turn it into an assigned service order. No slide presentations.\n\u2192 <a href=\"https:\/\/www.logicline.de\/en\/contact\"><strong>Schedule an introductory call<\/strong><\/a><\/p>\n\n<\/div>\n<h2>Which Rules in Machinery and Equipment Actually Hold True<\/h2>\nThe discussion surrounding predictive maintenance usually revolves around predictive models. In practice, a manageable number of rule-based triggers cover the vast majority of cases, and they have one key advantage: they are transparent. The head of service can explain to a customer why a service order was generated. With a model that outputs a risk score of 0.73, that becomes difficult.\n<figure><img fetchpriority=\"high\" decoding=\"async\" src=\"\/wp-content\/uploads\/2026\/08\/signal-zum-serviceauftrag-en.svg\" alt=\"Process from machine signal to service order: Telemetry and meter readings are fed into a routine check with four categories, which results in an assigned service order with an SLA time, machine history, and spare part recommendation.\" width=\"1600\" height=\"760\" \/><figcaption>The bottleneck is in the middle: It is the rule that decides whether a measured value results in a service order.<\/figcaption><\/figure>\nFour categories cover the full range of needs that mass-production manufacturers have in the field:\n<table>\n<thead>\n<tr>\n<th>Rule Category<\/th>\n<th>Trigger<\/th>\n<th>Typical Use<\/th>\n<th>What is created<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td>Threshold<\/td>\n<td>The measured value exceeds a limit<\/td>\n<td>Clearly defined limits, such as pressure or fill level<\/td>\n<td>Notification or service order, depending on the severity<\/td>\n<\/tr>\n<tr>\n<td>Persistent Condition<\/td>\n<td>The condition persists over a period of time<\/td>\n<td>Temperature, vibration, current consumption<\/td>\n<td>Service order without a dummy ticket for brief spikes<\/td>\n<\/tr>\n<tr>\n<td>Downtime and availability<\/td>\n<td>System does not respond or is down unexpectedly<\/td>\n<td>Availability guarantees in the contract<\/td>\n<td>Service order with SLA timer starting immediately<\/td>\n<\/tr>\n<tr>\n<td>Operating Hours and Cycles<\/td>\n<td>Counter reading reaches a maintenance point<\/td>\n<td>Wear parts, scheduleable maintenance<\/td>\n<td>Scheduled operation plus spare parts requirements<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\nThe fourth category is often underestimated. Operating hours and cycles have long been available in many control systems and can be retrieved without the need for a sensor project. They enable planned maintenance rather than emergency repairs, and they serve as a starting point for manufacturers with a technically heterogeneous fleet. Those who also explore usage-based models will find the commercial side of these same meter readings in our article on <a href=\"https:\/\/www.logicline.de\/en\/pay-per-use-mechanical-engineering-equipment-as-a-service\">pay-per-use and equipment-as-a-service<\/a>.\n\nThe effort involved does not scale with the number of machines, but rather with the number of machine types. A single rule per series applies to the entire fleet, making this approach cost-effective for mass-production manufacturers with a large installed base.\n<h3>Who defines and maintains the rules<\/h3>\nRules are rarely created at the IT desk. The development team knows the thresholds, the service team knows the error patterns, and the inside sales staff knows the contractual commitments. A small, cross-functional team has proven effective: rather than drafting a catalog that applies to all systems, it defines and implements a handful of triggers for each product series.\n\nThis includes a set inspection schedule. After the first few weeks, it becomes clear which rules trigger too often and which take effect too late, and both can only be adjusted if someone compares the triggered tasks with the actual findings. That\u2019s why the technician should be able to briefly note on the work order whether the trigger was justified. This feedback is the only reliable basis for fine-tuning thresholds and time windows, and it also serves as training data later on, should predictive models be added.\n<h2>What Needs to Happen in the Service Process<\/h2>\nAs soon as a rule is triggered, the part that decides on the benefit begins. The incident must be created as a service request in the system your service organization already uses, with everything that goes along with it: assignment to a resource with the appropriate qualifications, scheduling in the work plan, the SLA timer running against the contractual commitment, and an escalation path if the deadline is missed.\n\nThis requires a connection between three worlds that are often separated in many companies: machine data, master data on the installed base, and the service processes themselves. This is exactly where logicline comes in: signal analysis, the machine file, and service processing are integrated into a single, end-to-end workflow in Salesforce, rather than being three separate tools connected by interfaces. The article on <a href=\"https:\/\/www.logicline.de\/en\/making-iot-data-usable-in-the-service-portal\">IoT data in the service portal<\/a> describes how the data side of this chain works.\n\nIt is important to differentiate between them. Not every trigger justifies taking action. Some of the signals serve as a reminder for the system\u2014one that will be visible during the next scheduled maintenance visit\u2014while others generate a task with a deadline. If you fail to make this distinction, you will either create a flood of false alarms or miss a critical issue.\n\nThe opposite scenario to the one described above looks like this: The same temperature spike triggers no action. If the condition persists for ten minutes, a service order is automatically generated that already includes the system\u2019s service history, the contract status, and a spare parts recommendation. Scheduling sees it in the work plan in the morning\u2014not as a tile that disappeared overnight.\n<h3>What the prerequisite for this is<\/h3>\nA rule can only be as precise as the underlying master data. If it is unclear which machine, with which configuration, is operating at which customer site under which contract, no trigger can be properly assigned and no SLA can be correctly initiated. That is why the structured installed base is the first step, not the second. The article on <a href=\"https:\/\/www.logicline.de\/en\/predictive-maintenance-data-foundation-machinery-and-equipment-manufacturers\">the data foundation for predictive maintenance<\/a> addresses this level; here, we\u2019re focusing on the level that follows.\n<h2>Approach: Connect first, then decide<\/h2>\nIn <strong>the \u201cDigitize \u2192 Connect \u2192 Decide \u2192 Automate\u201d model<\/strong>, rule-based signal processing occurs at stage two. Stage one provides the structured master data, without which no trigger can be uniquely identified. An alarm is only as valuable as the process into which it feeds.\n\nAt level three, the question shifts. The triggered incident becomes a triaged case: How urgent is it compared to the other open work orders? Given the history of this system, what is the most likely cause? Which part should be shipped along with it? <strong>Service Decision Intelligence (SDI)<\/strong> performs this assessment as an intelligence layer over machine, work order, and document data.\n\nTwo features are crucial for machinery and equipment manufacturers in this context. First, the source citation: Each recommendation specifies the data and documents on which it is based, which can be verified by customers and in the event of a warranty claim. Second, data sovereignty: SDI runs on the customer\u2019s infrastructure, is EU-compliant, and is independent of the language model used. Operational data from the installed base is part of the core business and does not leave the company.\n<h2>Next step<\/h2>\nIf your machines are generating data but the service department has not been making much use of it so far, the cause rarely lies with the sensors. It lies in the lack of control logic and in the separation between the alarm channel and the service processes. Both of these gaps can be closed step by step, starting with a single product series and a handful of triggers.\n\nTwo pragmatic approaches:\n<ul>\n \t<li><strong><a href=\"https:\/\/www.logicline.de\/en\/services\/installed-base-assessment\">Installed Base Assessment<\/a><\/strong> \u2013 when it is unclear which machine is running in which configuration under which contract. Without this mapping, no rule will apply.<\/li>\n \t<li><strong><a href=\"https:\/\/www.logicline.de\/en\/contact\">introductory call<\/a><\/strong> \u2013 once the master data is in place and you want to define the initial triggers, including deciding which product series to start with.<\/li>\n<\/ul>\nTo start with, choose the production line that had the most unscheduled operations last year. That&#8217;s where the difference between an emergency and a scheduled appointment is most easily measurable.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t<section data-dce-background-color=\"#00000000\" class=\"elementor-element elementor-element-c254285 e-flex e-con-boxed e-con e-parent\" data-id=\"c254285\" data-element_type=\"container\" data-e-type=\"container\" data-settings=\"{&quot;background_background&quot;:&quot;classic&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t<div data-dce-background-color=\"#FFFFFF\" class=\"elementor-element elementor-element-55aff5e e-flex e-con-boxed e-con e-child\" data-id=\"55aff5e\" data-element_type=\"container\" data-e-type=\"container\" data-settings=\"{&quot;background_background&quot;:&quot;classic&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t<div class=\"elementor-element elementor-element-48de356 e-con-full e-flex e-con e-child\" data-id=\"48de356\" data-element_type=\"container\" data-e-type=\"container\">\n\t\t<div class=\"elementor-element elementor-element-4b7d6a6 e-flex e-con-boxed e-con e-child\" data-id=\"4b7d6a6\" data-element_type=\"container\" data-e-type=\"container\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-0fa94fe elementor-widget elementor-widget-heading\" data-id=\"0fa94fe\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<h2 class=\"elementor-heading-title elementor-size-default\">FAQs<\/h2>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-ed3a7f2 elementor-widget elementor-widget-n-accordion\" data-id=\"ed3a7f2\" data-element_type=\"widget\" data-e-type=\"widget\" data-settings=\"{&quot;default_state&quot;:&quot;expanded&quot;,&quot;max_items_expended&quot;:&quot;one&quot;,&quot;n_accordion_animation_duration&quot;:{&quot;unit&quot;:&quot;ms&quot;,&quot;size&quot;:400,&quot;sizes&quot;:[]}}\" data-widget_type=\"nested-accordion.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<div class=\"e-n-accordion\" aria-label=\"Accordion. Open links with Enter or Space, close with Escape, and navigate with Arrow Keys\">\n\t\t\t\t\t\t<details id=\"e-n-accordion-item-2480\" class=\"e-n-accordion-item\" open>\n\t\t\t\t<summary class=\"e-n-accordion-item-title\" data-accordion-index=\"1\" tabindex=\"0\" aria-expanded=\"true\" aria-controls=\"e-n-accordion-item-2480\" >\n\t\t\t\t\t<span class='e-n-accordion-item-title-header'><h4 class=\"e-n-accordion-item-title-text\"> How is machine data converted into a service order? <\/h4><\/span>\n\t\t\t\t\t\t\t<span class='e-n-accordion-item-title-icon'>\n\t\t\t<span class='e-opened' ><svg aria-hidden=\"true\" class=\"e-font-icon-svg e-fas-minus\" viewBox=\"0 0 448 512\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M416 208H32c-17.67 0-32 14.33-32 32v32c0 17.67 14.33 32 32 32h384c17.67 0 32-14.33 32-32v-32c0-17.67-14.33-32-32-32z\"><\/path><\/svg><\/span>\n\t\t\t<span class='e-closed'><svg aria-hidden=\"true\" class=\"e-font-icon-svg e-fas-plus\" viewBox=\"0 0 448 512\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M416 208H272V64c0-17.67-14.33-32-32-32h-32c-17.67 0-32 14.33-32 32v144H32c-17.67 0-32 14.33-32 32v32c0 17.67 14.33 32 32 32h144v144c0 17.67 14.33 32 32 32h32c17.67 0 32-14.33 32-32V304h144c17.67 0 32-14.33 32-32v-32c0-17.67-14.33-32-32-32z\"><\/path><\/svg><\/span>\n\t\t<\/span>\n\n\t\t\t\t\t\t<\/summary>\n\t\t\t\t<div role=\"region\" aria-labelledby=\"e-n-accordion-item-2480\" class=\"elementor-element elementor-element-8046bac e-flex e-con-boxed e-con e-child\" data-id=\"8046bac\" data-element_type=\"container\" data-e-type=\"container\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-e3a4f16 elementor-widget elementor-widget-text-editor\" data-id=\"e3a4f16\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>In three steps: Telemetry data or a meter reading triggers a rule check that determines what constitutes an incident. If the rule is triggered, a service order is automatically created in the service system, complete with assignment, schedule, and an active SLA timer. This order includes the machine context: service history, configuration, contract status, and spare parts procurement. If any of the three steps is missing, the measured value remains merely a reading with no effect.   <\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/details>\n\t\t\t\t\t\t<details id=\"e-n-accordion-item-2481\" class=\"e-n-accordion-item\" >\n\t\t\t\t<summary class=\"e-n-accordion-item-title\" data-accordion-index=\"2\" tabindex=\"-1\" aria-expanded=\"false\" aria-controls=\"e-n-accordion-item-2481\" >\n\t\t\t\t\t<span class='e-n-accordion-item-title-header'><h4 class=\"e-n-accordion-item-title-text\"> How can you prevent false alarms in condition monitoring? <\/h4><\/span>\n\t\t\t\t\t\t\t<span class='e-n-accordion-item-title-icon'>\n\t\t\t<span class='e-opened' ><svg aria-hidden=\"true\" class=\"e-font-icon-svg e-fas-minus\" viewBox=\"0 0 448 512\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M416 208H32c-17.67 0-32 14.33-32 32v32c0 17.67 14.33 32 32 32h384c17.67 0 32-14.33 32-32v-32c0-17.67-14.33-32-32-32z\"><\/path><\/svg><\/span>\n\t\t\t<span class='e-closed'><svg aria-hidden=\"true\" class=\"e-font-icon-svg e-fas-plus\" viewBox=\"0 0 448 512\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M416 208H272V64c0-17.67-14.33-32-32-32h-32c-17.67 0-32 14.33-32 32v144H32c-17.67 0-32 14.33-32 32v32c0 17.67 14.33 32 32 32h144v144c0 17.67 14.33 32 32 32h32c17.67 0 32-14.33 32-32V304h144c17.67 0 32-14.33 32-32v-32c0-17.67-14.33-32-32-32z\"><\/path><\/svg><\/span>\n\t\t<\/span>\n\n\t\t\t\t\t\t<\/summary>\n\t\t\t\t<div role=\"region\" aria-labelledby=\"e-n-accordion-item-2481\" class=\"elementor-element elementor-element-5414f61 e-con-full e-flex e-con e-child\" data-id=\"5414f61\" data-element_type=\"container\" data-e-type=\"container\">\n\t\t\t\t<div class=\"elementor-element elementor-element-1e5ed46 elementor-widget elementor-widget-text-editor\" data-id=\"1e5ed46\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>On persistence conditions rather than pure threshold values. A threshold value alone reacts to startup processes, load changes, and environmental influences. Only the condition \u201cvalue exceeded continuously over a defined time window while the system is running\u201d distinguishes the event from a mere fluctuation. Additionally, a tiered approach helps: Not every trigger needs to generate a service order; some are sufficient as a warning on the system.   <\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/details>\n\t\t\t\t\t\t<details id=\"e-n-accordion-item-2482\" class=\"e-n-accordion-item\" >\n\t\t\t\t<summary class=\"e-n-accordion-item-title\" data-accordion-index=\"3\" tabindex=\"-1\" aria-expanded=\"false\" aria-controls=\"e-n-accordion-item-2482\" >\n\t\t\t\t\t<span class='e-n-accordion-item-title-header'><h4 class=\"e-n-accordion-item-title-text\"> Is additional sensor technology absolutely necessary for predictive maintenance? <\/h4><\/span>\n\t\t\t\t\t\t\t<span class='e-n-accordion-item-title-icon'>\n\t\t\t<span class='e-opened' ><svg aria-hidden=\"true\" class=\"e-font-icon-svg e-fas-minus\" viewBox=\"0 0 448 512\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M416 208H32c-17.67 0-32 14.33-32 32v32c0 17.67 14.33 32 32 32h384c17.67 0 32-14.33 32-32v-32c0-17.67-14.33-32-32-32z\"><\/path><\/svg><\/span>\n\t\t\t<span class='e-closed'><svg aria-hidden=\"true\" class=\"e-font-icon-svg e-fas-plus\" viewBox=\"0 0 448 512\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M416 208H272V64c0-17.67-14.33-32-32-32h-32c-17.67 0-32 14.33-32 32v144H32c-17.67 0-32 14.33-32 32v32c0 17.67 14.33 32 32 32h144v144c0 17.67 14.33 32 32 32h32c17.67 0 32-14.33 32-32V304h144c17.67 0 32-14.33 32-32v-32c0-17.67-14.33-32-32-32z\"><\/path><\/svg><\/span>\n\t\t<\/span>\n\n\t\t\t\t\t\t<\/summary>\n\t\t\t\t<div role=\"region\" aria-labelledby=\"e-n-accordion-item-2482\" class=\"elementor-element elementor-element-cca8e05 e-con-full e-flex e-con e-child\" data-id=\"cca8e05\" data-element_type=\"container\" data-e-type=\"container\">\n\t\t\t\t<div class=\"elementor-element elementor-element-7504c5c elementor-widget elementor-widget-text-editor\" data-id=\"7504c5c\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>Usually not for getting started. Many control systems already have operating hours and cycle counters, which generate predictable maintenance triggers\u2014including spare part requirements\u2014without the need for a retrofit project. Additional sensors are worthwhile in cases where wear and tear does not correlate with service life and a failure is costly. The order of priority is: first use the existing signals, then retrofit selectively.   <\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/details>\n\t\t\t\t\t\t<details id=\"e-n-accordion-item-2483\" class=\"e-n-accordion-item\" >\n\t\t\t\t<summary class=\"e-n-accordion-item-title\" data-accordion-index=\"4\" tabindex=\"-1\" aria-expanded=\"false\" aria-controls=\"e-n-accordion-item-2483\" >\n\t\t\t\t\t<span class='e-n-accordion-item-title-header'><h4 class=\"e-n-accordion-item-title-text\"> Where is the line between rules and predictive models? <\/h4><\/span>\n\t\t\t\t\t\t\t<span class='e-n-accordion-item-title-icon'>\n\t\t\t<span class='e-opened' ><svg aria-hidden=\"true\" class=\"e-font-icon-svg e-fas-minus\" viewBox=\"0 0 448 512\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M416 208H32c-17.67 0-32 14.33-32 32v32c0 17.67 14.33 32 32 32h384c17.67 0 32-14.33 32-32v-32c0-17.67-14.33-32-32-32z\"><\/path><\/svg><\/span>\n\t\t\t<span class='e-closed'><svg aria-hidden=\"true\" class=\"e-font-icon-svg e-fas-plus\" viewBox=\"0 0 448 512\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M416 208H272V64c0-17.67-14.33-32-32-32h-32c-17.67 0-32 14.33-32 32v144H32c-17.67 0-32 14.33-32 32v32c0 17.67 14.33 32 32 32h144v144c0 17.67 14.33 32 32 32h32c17.67 0 32-14.33 32-32V304h144c17.67 0 32-14.33 32-32v-32c0-17.67-14.33-32-32-32z\"><\/path><\/svg><\/span>\n\t\t<\/span>\n\n\t\t\t\t\t\t<\/summary>\n\t\t\t\t<div role=\"region\" aria-labelledby=\"e-n-accordion-item-2483\" class=\"elementor-element elementor-element-4fc6b46 e-con-full e-flex e-con e-child\" data-id=\"4fc6b46\" data-element_type=\"container\" data-e-type=\"container\">\n\t\t\t\t<div class=\"elementor-element elementor-element-79f5387 elementor-widget elementor-widget-text-editor\" data-id=\"79f5387\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>Rules cover the majority of recurring cases and are transparent, which is important when dealing with customers and in warranty claims. Predictive models are useful when a fault pattern arises from the interaction of multiple signals and no one can specify a meaningful threshold. In practice, you start with rules because they take effect more quickly and provide the data quality that a model will need later anyway.  <\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/details>\n\t\t\t\t\t\t<details id=\"e-n-accordion-item-2484\" class=\"e-n-accordion-item\" >\n\t\t\t\t<summary class=\"e-n-accordion-item-title\" data-accordion-index=\"5\" tabindex=\"-1\" aria-expanded=\"false\" aria-controls=\"e-n-accordion-item-2484\" >\n\t\t\t\t\t<span class='e-n-accordion-item-title-header'><h4 class=\"e-n-accordion-item-title-text\"> What needs to be in place before the first rule goes live? <\/h4><\/span>\n\t\t\t\t\t\t\t<span class='e-n-accordion-item-title-icon'>\n\t\t\t<span class='e-opened' ><svg aria-hidden=\"true\" class=\"e-font-icon-svg e-fas-minus\" viewBox=\"0 0 448 512\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M416 208H32c-17.67 0-32 14.33-32 32v32c0 17.67 14.33 32 32 32h384c17.67 0 32-14.33 32-32v-32c0-17.67-14.33-32-32-32z\"><\/path><\/svg><\/span>\n\t\t\t<span class='e-closed'><svg aria-hidden=\"true\" class=\"e-font-icon-svg e-fas-plus\" viewBox=\"0 0 448 512\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M416 208H272V64c0-17.67-14.33-32-32-32h-32c-17.67 0-32 14.33-32 32v144H32c-17.67 0-32 14.33-32 32v32c0 17.67 14.33 32 32 32h144v144c0 17.67 14.33 32 32 32h32c17.67 0 32-14.33 32-32V304h144c17.67 0 32-14.33 32-32v-32c0-17.67-14.33-32-32-32z\"><\/path><\/svg><\/span>\n\t\t<\/span>\n\n\t\t\t\t\t\t<\/summary>\n\t\t\t\t<div role=\"region\" aria-labelledby=\"e-n-accordion-item-2484\" class=\"elementor-element elementor-element-0e62110 e-con-full e-flex e-con e-child\" data-id=\"0e62110\" data-element_type=\"container\" data-e-type=\"container\">\n\t\t\t\t<div class=\"elementor-element elementor-element-60b8663 elementor-widget elementor-widget-text-editor\" data-id=\"60b8663\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>A clear mapping of the asset: which machine, which configuration, which customer, which contract. Without this master data, a trigger cannot be addressed correctly, nor can the appropriate SLA be initiated. In addition, the team must decide which triggers generate a service order and which merely notify the team, so that the team trusts the channel from the very beginning.  <\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/details>\n\t\t\t\t\t\t<details id=\"e-n-accordion-item-2485\" class=\"e-n-accordion-item\" >\n\t\t\t\t<summary class=\"e-n-accordion-item-title\" data-accordion-index=\"6\" tabindex=\"-1\" aria-expanded=\"false\" aria-controls=\"e-n-accordion-item-2485\" >\n\t\t\t\t\t<span class='e-n-accordion-item-title-header'><h4 class=\"e-n-accordion-item-title-text\"> What role does Service Decision Intelligence play in this? <\/h4><\/span>\n\t\t\t\t\t\t\t<span class='e-n-accordion-item-title-icon'>\n\t\t\t<span class='e-opened' ><svg aria-hidden=\"true\" class=\"e-font-icon-svg e-fas-minus\" viewBox=\"0 0 448 512\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M416 208H32c-17.67 0-32 14.33-32 32v32c0 17.67 14.33 32 32 32h384c17.67 0 32-14.33 32-32v-32c0-17.67-14.33-32-32-32z\"><\/path><\/svg><\/span>\n\t\t\t<span class='e-closed'><svg aria-hidden=\"true\" class=\"e-font-icon-svg e-fas-plus\" viewBox=\"0 0 448 512\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><path d=\"M416 208H272V64c0-17.67-14.33-32-32-32h-32c-17.67 0-32 14.33-32 32v144H32c-17.67 0-32 14.33-32 32v32c0 17.67 14.33 32 32 32h144v144c0 17.67 14.33 32 32 32h32c17.67 0 32-14.33 32-32V304h144c17.67 0 32-14.33 32-32v-32c0-17.67-14.33-32-32-32z\"><\/path><\/svg><\/span>\n\t\t<\/span>\n\n\t\t\t\t\t\t<\/summary>\n\t\t\t\t<div role=\"region\" aria-labelledby=\"e-n-accordion-item-2485\" class=\"elementor-element elementor-element-b64a09a e-con-full e-flex e-con e-child\" data-id=\"b64a09a\" data-element_type=\"container\" data-e-type=\"container\">\n\t\t\t\t<div class=\"elementor-element elementor-element-857ea34 elementor-widget elementor-widget-text-editor\" data-id=\"857ea34\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>The rule decides whether an incident has occurred. Service Decision Intelligence (SDI) evaluates it: urgency compared to other open work orders, probable cause based on the equipment history, and the appropriate spare part. Every recommendation includes a source citation, and SDI runs on the customer\u2019s infrastructure, is EU-compliant, and is independent of the language model. This ensures that operational data for the installed base remains in-house.   <\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/details>\n\t\t\t\t\t<\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/section>\n\t\t\t\t<\/div>\n\t\t","protected":false},"excerpt":{"rendered":"<p>At 2:14 a.m., a unit in the field reports a temperature spike. The value is sent to an IoT application, where a red tile appears. No one is on call for this tile because it isn&#8217;t a service process. By morning, the value is back in the green range, the tile disappears, and no one [&hellip;]<\/p>\n","protected":false},"author":5,"featured_media":42598,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"inline_featured_image":false,"footnotes":""},"categories":[43051],"tags":[],"class_list":["post-42603","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-iot-installed-base-en"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.logicline.de\/en\/wp-json\/wp\/v2\/posts\/42603","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.logicline.de\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.logicline.de\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.logicline.de\/en\/wp-json\/wp\/v2\/users\/5"}],"replies":[{"embeddable":true,"href":"https:\/\/www.logicline.de\/en\/wp-json\/wp\/v2\/comments?post=42603"}],"version-history":[{"count":5,"href":"https:\/\/www.logicline.de\/en\/wp-json\/wp\/v2\/posts\/42603\/revisions"}],"predecessor-version":[{"id":42608,"href":"https:\/\/www.logicline.de\/en\/wp-json\/wp\/v2\/posts\/42603\/revisions\/42608"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.logicline.de\/en\/wp-json\/wp\/v2\/media\/42598"}],"wp:attachment":[{"href":"https:\/\/www.logicline.de\/en\/wp-json\/wp\/v2\/media?parent=42603"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.logicline.de\/en\/wp-json\/wp\/v2\/categories?post=42603"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.logicline.de\/en\/wp-json\/wp\/v2\/tags?post=42603"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}