{"id":3372,"date":"2023-08-31T10:59:22","date_gmt":"2023-08-31T10:59:22","guid":{"rendered":"https:\/\/neuraldesigner.com\/blog\/auto-associative-neural-networks\/"},"modified":"2025-09-10T17:21:41","modified_gmt":"2025-09-10T15:21:41","slug":"aanns-predictive-maintenance-guide","status":"publish","type":"blog","link":"https:\/\/www.neuraldesigner.com\/blog\/aanns-predictive-maintenance-guide\/","title":{"rendered":"Auto-associative neural networks (AANNs) for predictive maintenance"},"content":{"rendered":"<section>\n<section>\n<h2>1. Introduction<\/h2>\n<p>Predictive maintenance uses data analysis and machine learning techniques to identify potential equipment failures before they occur.<\/p>\n<p>It involves collecting and analyzing data from equipment and systems to predict when maintenance will be required and to take action to prevent equipment failure and downtime.<\/p>\n<p>For instance, Auto-Associative Neural Networks (AANNs) are a powerful tool that we can use to identify abnormal states in data and help us with predictive maintenance.<\/p>\n<p>Predictive maintenance is essential for several reasons, including:<\/p>\n<ul>\n<li><strong>Reduced downtime:<\/strong> Predictive maintenance can identify and address potential issues before they become critical failures. This helps avoid unexpected downtime.<\/li>\n<li><strong>Increased safety:<\/strong> Predictive maintenance can prevent accidents and injuries caused by malfunctioning equipment.<\/li>\n<li><strong>Cost savings:<\/strong> Predictive maintenance can prevent minor problems from becoming significant failures, which can be costly to repair.<\/li>\n<li><strong>Better asset management:<\/strong> Predictive maintenance can help organizations make informed decisions about maintenance schedules, replacement timelines, and asset retirement.<\/li>\n<li><strong>Improved equipment life:<\/strong> Predictive maintenance can help prevent premature equipment failure.<\/li>\n<\/ul>\n<p>Predictive maintenance is a complex problem that requires analyzing vast amounts of data from different sources, including equipment sensors, maintenance records, and other relevant factors such as environmental conditions and operating parameters.<\/p>\n<p>However, more than historical data is needed to capture the full range of factors that influence equipment performance, making it difficult to build accurate predictive models.<\/p>\n<p>Traditional statistical and machine learning techniques may need to be more effective in identifying patterns and predicting failures in complex systems.<\/p>\n<p>Unlike conventional statistical models, which rely on historical data, Auto-Associative Neural Networks (AANNs) can learn underlying patterns and structures and use that information to detect significant deviations from what is considered &#8220;normal.&#8221;<img fetchpriority=\"high\" decoding=\"async\" src=\"https:\/\/www.neuraldesigner.com\/images\/auto-associative_neural_networks.webp\" alt=\"An example of an Auto-Associative Neural Networks (AANNs) for predictive maintenance.\" width=\"900\" height=\"750\" \/><\/p>\n<p>To sum up, anomaly detection involves identifying rare or unexpected events in data that deviate significantly from what we consider normal.<\/p>\n<\/section>\n<h3>Contents<\/h3>\n<ol>\n<li><a href=\"#Architecture\">Architecture<\/a>.<\/li>\n<li><a href=\"#Training\">Training<\/a>.<\/li>\n<li><a href=\"#Deployment\">Deployment<\/a>.<\/li>\n<li><a href=\"#Use Cases\">Other use cases<\/a>.<\/li>\n<li><a href=\"#References\">References<\/a>.<\/li>\n<\/ol>\n<\/section>\n<section>\n<h2>2. Architecture<\/h2>\n<p>The <a href=\"https:\/\/www.neuraldesigner.com\/learning\/tutorials\/neural-network#NetworkArchitecture\">architecture<\/a> of this network consists of several layers: the input layer, the mapping layer, the bottleneck layer, and the decoding or de-mapping layer.<\/p>\n<h3><strong>Input layer<\/strong><\/h3>\n<p>It is the first layer of the network and is responsible for receiving incoming data.<\/p>\n<p>Each neuron in this layer represents a variable or characteristic in the input data set.<\/p>\n<p>The activation function used in this layer depends on the data type being used.<\/p>\n<p>For example, the <a href=\"https:\/\/www.neuraldesigner.com\/learning\/tutorials\/neural-network#LinearActivationFunction\">linear activation function<\/a> is usually used for numerical data.<\/p>\n<h3><strong>Mapping layer<\/strong><\/h3>\n<p>It is the second layer of the network.<\/p>\n<p>Specifically, this layer maps the input data to a latent representation in a new dimension.<\/p>\n<p>Moreover, each neuron in this layer receives inputs from all neurons in the input layer and, consequently, emits a unique output that represents a linear combination of the inputs received.<\/p>\n<p>Furthermore, the activation function used in this layer is usually nonlinear, such as the <a href=\"https:\/\/www.neuraldesigner.com\/learning\/tutorials\/neural-network#HyperbolicTangentActivationFunction\">sigmoid function<\/a> or the <a href=\"https:\/\/www.neuraldesigner.com\/learning\/tutorials\/neural-network#RectifiedLinearActivationFunction\">ReLU function<\/a>.<\/p>\n<h3><strong>Bottleneck layer<\/strong><\/h3>\n<p>It is the third layer of the network and is responsible for reducing the dimensionality of the representation.<\/p>\n<p>This layer has fewer neurons than the input and mapping layers.<\/p>\n<p>Therefore, it acts as a bottleneck that forces the network to learn a more compact representation of the data.<\/p>\n<p>The activation function used in this layer is usually the <a href=\"https:\/\/www.neuraldesigner.com\/learning\/tutorials\/neural-network#LinearActivationFunction\">linear function<\/a>, regardless of the data used.<\/p>\n<h3><strong>Decoding layer<\/strong><\/h3>\n<p>After passing through the bottleneck layer, the representation propagates to the decoding layer, a reverse copy of the mapping layer.<\/p>\n<p>The decoding layer is responsible for reconstructing the original input from the representation.<\/p>\n<p>Therefore, the activation function used in this layer must be the same as in the mapping layer.<\/p>\n<p>In summary, the input layer receives the input data, the mapping layer learns a latent representation of the data, the bottleneck layer reduces the dimensionality of the latent representation, and the decoding layer reconstructs the original input from the latent representation.<\/p>\n<\/section>\n<section>\n<h2>3. Training<\/h2>\n<p>Using auto-associative neural networks (AANNs) for anomaly detection in predictive maintenance starts with <a href=\"https:\/\/www.neuraldesigner.com\/learning\/tutorials\/training-strategy\">training<\/a> the network on a &#8220;normal&#8221; data set, representing the typical or expected behavior.<\/p>\n<p>The network then learns to compress this data into a lower-dimensional representation that captures the essential features of the data.<\/p>\n<p>The network&#8217;s hidden layer, which contains this compressed representation, is then used to reconstruct the original data.<\/p>\n<p>However, strange patterns in the input data will not fit well with the compact representation learned by the network.<\/p>\n<p>Therefore, if the network is used to reconstruct the anomalous input data, the reconstructed output will significantly differ from the original.<\/p>\n<p>This difference between the original and rebuilt data can be quantified using a measure of distance or error, such as <a href=\"https:\/\/www.neuraldesigner.com\/learning\/tutorials\/training-strategy#MeanSquaredError\">Mean Squared Error (MSE)<\/a>:<\/p>\n<p>[<br \/>\nbegin{eqnarray}<br \/>\ntext{MSE} = frac{1}{n} sum_{i = 1}^{n} left( y_i &#8211; tilde{y}_{i} right)^2<br \/>\nend{eqnarray}<br \/>\n]<br \/>\nOr Euclidean distance:<br \/>\n[<br \/>\nbegin{eqnarray}<br \/>\nd(x,y) = sqrt{sum_{i = 1}^{n} left| y_i &#8211; x_i right|^2 }<br \/>\nend{eqnarray}<br \/>\n]<\/p>\n<p>Once we have trained the model, the distances between the encoded input data and the original input data are saved. Then, we can use these distances during deployment.<\/p>\n<p><img decoding=\"async\" src=\"https:\/\/www.neuraldesigner.com\/images\/Samples_distances_distribution.webp\" alt=\"\" width=\"500\" \/><\/p>\n<p>A threshold for acceptable error is defined for the normal data. Any new data that results in a reconstruction error exceeding the established threshold will be identified as abnormal.<\/p>\n<\/section>\n<section>\n<h2>4. Deployment<\/h2>\n<p>The <a href=\"https:\/\/www.neuraldesigner.com\/learning\/tutorials\/model-deployment\">deployment<\/a> stage involves passing new data through the trained model, encoding it into a new representation, and comparing the distances between the encoded input data and the original input data to determine whether it is within an acceptable range.<\/p>\n<p><img decoding=\"async\" src=\"https:\/\/www.neuraldesigner.com\/images\/Sample_distance_box_plot.webp\" alt=\"\" width=\"500\" \/><\/p>\n<p>If the distance exceeds the established threshold, the input data is considered abnormal, and further investigation may be required.<\/p>\n<p>The deployment process also involves monitoring the model&#8217;s performance over time and making necessary updates and adjustments to ensure that it continues to make accurate predictions on new data.<\/p>\n<\/section>\n<section>\n<h2>5. Other use cases<\/h2>\n<p>Auto-associative Neural Networks (AANNs) aren&#8217;t used only for predictive maintenance; they have a wide range of applications in various industries.<\/p>\n<ul>\n<li><strong>Cybersecurity: <\/strong>They can be used to detect anomalous patterns in network traffic and identify potential security breaches.<\/li>\n<li><strong>Healthcare:<\/strong> They can be used to detect anomalous patient behavior, such as irregular heartbeats or abnormal medical readings.<\/li>\n<li><strong>Finance:<\/strong> They can be used for fraud detection, identifying unusual patterns in financial transactions, and flagging them for further investigation.<\/li>\n<\/ul>\n<p>These networks&#8217; potential uses are vast and varied, making them valuable tools.<\/p>\n<\/section>\n<section>\n<h2>6. References<\/h2>\n<ul>\n<li>Hinton, G. E., &amp; Salakhutdinov, R. R. (2006). Reducing the dimensionality of data with neural networks. <a href=\"https:\/\/www.science.org\/doi\/10.1126\/science.1127647\">Science<\/a>\u00a0 313(5786), 504-507.<\/li>\n<\/ul>\n<h2>Related posts<\/h2>\n<\/section>\n","protected":false},"author":11,"featured_media":2634,"template":"","categories":[],"tags":[36],"class_list":["post-3372","blog","type-blog","status-publish","has-post-thumbnail","hentry","tag-tutorials"],"acf":[],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v26.4 - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>Auto-associative neural networks (AANNs) for predictive maintenance<\/title>\n<meta name=\"description\" content=\"Explore Auto-Associative Neural Networks (AANNs) for predictive maintenance and learn about architecture, training and deployment.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.neuraldesigner.com\/blog\/aanns-predictive-maintenance-guide\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Auto-associative neural networks (AANNs) for predictive maintenance\" \/>\n<meta property=\"og:description\" content=\"Explore Auto-Associative Neural Networks (AANNs) for predictive maintenance and learn about architecture, training and deployment.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.neuraldesigner.com\/blog\/aanns-predictive-maintenance-guide\/\" \/>\n<meta property=\"og:site_name\" content=\"Neural Designer\" \/>\n<meta property=\"article:modified_time\" content=\"2025-09-10T15:21:41+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/06\/auto-associative_neural_networks.webp\" \/>\n\t<meta property=\"og:image:width\" content=\"1215\" \/>\n\t<meta property=\"og:image:height\" content=\"750\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/webp\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:site\" content=\"@NeuralDesigner\" \/>\n<meta name=\"twitter:label1\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data1\" content=\"6 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/www.neuraldesigner.com\/blog\/aanns-predictive-maintenance-guide\/\",\"url\":\"https:\/\/www.neuraldesigner.com\/blog\/aanns-predictive-maintenance-guide\/\",\"name\":\"Auto-associative neural networks (AANNs) for predictive maintenance\",\"isPartOf\":{\"@id\":\"https:\/\/www.neuraldesigner.com\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/www.neuraldesigner.com\/blog\/aanns-predictive-maintenance-guide\/#primaryimage\"},\"image\":{\"@id\":\"https:\/\/www.neuraldesigner.com\/blog\/aanns-predictive-maintenance-guide\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/06\/auto-associative_neural_networks.webp\",\"datePublished\":\"2023-08-31T10:59:22+00:00\",\"dateModified\":\"2025-09-10T15:21:41+00:00\",\"description\":\"Explore Auto-Associative Neural Networks (AANNs) for predictive maintenance and learn about architecture, training and deployment.\",\"breadcrumb\":{\"@id\":\"https:\/\/www.neuraldesigner.com\/blog\/aanns-predictive-maintenance-guide\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/www.neuraldesigner.com\/blog\/aanns-predictive-maintenance-guide\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/www.neuraldesigner.com\/blog\/aanns-predictive-maintenance-guide\/#primaryimage\",\"url\":\"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/06\/auto-associative_neural_networks.webp\",\"contentUrl\":\"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/06\/auto-associative_neural_networks.webp\",\"width\":1215,\"height\":750},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/www.neuraldesigner.com\/blog\/aanns-predictive-maintenance-guide\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\/\/www.neuraldesigner.com\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Blog\",\"item\":\"https:\/\/www.neuraldesigner.com\/blog\/\"},{\"@type\":\"ListItem\",\"position\":3,\"name\":\"Auto-associative neural networks (AANNs) for predictive maintenance\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.neuraldesigner.com\/#website\",\"url\":\"https:\/\/www.neuraldesigner.com\/\",\"name\":\"Neural Designer\",\"description\":\"Explanable AI Platform\",\"publisher\":{\"@id\":\"https:\/\/www.neuraldesigner.com\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/www.neuraldesigner.com\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/www.neuraldesigner.com\/#organization\",\"name\":\"Neural Designer\",\"url\":\"https:\/\/www.neuraldesigner.com\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/www.neuraldesigner.com\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/05\/logo-neural-1.png\",\"contentUrl\":\"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/05\/logo-neural-1.png\",\"width\":1024,\"height\":223,\"caption\":\"Neural Designer\"},\"image\":{\"@id\":\"https:\/\/www.neuraldesigner.com\/#\/schema\/logo\/image\/\"},\"sameAs\":[\"https:\/\/x.com\/NeuralDesigner\",\"https:\/\/es.linkedin.com\/showcase\/neuraldesigner\/\"]}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Auto-associative neural networks (AANNs) for predictive maintenance","description":"Explore Auto-Associative Neural Networks (AANNs) for predictive maintenance and learn about architecture, training and deployment.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.neuraldesigner.com\/blog\/aanns-predictive-maintenance-guide\/","og_locale":"en_US","og_type":"article","og_title":"Auto-associative neural networks (AANNs) for predictive maintenance","og_description":"Explore Auto-Associative Neural Networks (AANNs) for predictive maintenance and learn about architecture, training and deployment.","og_url":"https:\/\/www.neuraldesigner.com\/blog\/aanns-predictive-maintenance-guide\/","og_site_name":"Neural Designer","article_modified_time":"2025-09-10T15:21:41+00:00","og_image":[{"width":1215,"height":750,"url":"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/06\/auto-associative_neural_networks.webp","type":"image\/webp"}],"twitter_card":"summary_large_image","twitter_site":"@NeuralDesigner","twitter_misc":{"Est. reading time":"6 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/www.neuraldesigner.com\/blog\/aanns-predictive-maintenance-guide\/","url":"https:\/\/www.neuraldesigner.com\/blog\/aanns-predictive-maintenance-guide\/","name":"Auto-associative neural networks (AANNs) for predictive maintenance","isPartOf":{"@id":"https:\/\/www.neuraldesigner.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/www.neuraldesigner.com\/blog\/aanns-predictive-maintenance-guide\/#primaryimage"},"image":{"@id":"https:\/\/www.neuraldesigner.com\/blog\/aanns-predictive-maintenance-guide\/#primaryimage"},"thumbnailUrl":"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/06\/auto-associative_neural_networks.webp","datePublished":"2023-08-31T10:59:22+00:00","dateModified":"2025-09-10T15:21:41+00:00","description":"Explore Auto-Associative Neural Networks (AANNs) for predictive maintenance and learn about architecture, training and deployment.","breadcrumb":{"@id":"https:\/\/www.neuraldesigner.com\/blog\/aanns-predictive-maintenance-guide\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.neuraldesigner.com\/blog\/aanns-predictive-maintenance-guide\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.neuraldesigner.com\/blog\/aanns-predictive-maintenance-guide\/#primaryimage","url":"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/06\/auto-associative_neural_networks.webp","contentUrl":"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/06\/auto-associative_neural_networks.webp","width":1215,"height":750},{"@type":"BreadcrumbList","@id":"https:\/\/www.neuraldesigner.com\/blog\/aanns-predictive-maintenance-guide\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/www.neuraldesigner.com\/"},{"@type":"ListItem","position":2,"name":"Blog","item":"https:\/\/www.neuraldesigner.com\/blog\/"},{"@type":"ListItem","position":3,"name":"Auto-associative neural networks (AANNs) for predictive maintenance"}]},{"@type":"WebSite","@id":"https:\/\/www.neuraldesigner.com\/#website","url":"https:\/\/www.neuraldesigner.com\/","name":"Neural Designer","description":"Explanable AI Platform","publisher":{"@id":"https:\/\/www.neuraldesigner.com\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.neuraldesigner.com\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/www.neuraldesigner.com\/#organization","name":"Neural Designer","url":"https:\/\/www.neuraldesigner.com\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.neuraldesigner.com\/#\/schema\/logo\/image\/","url":"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/05\/logo-neural-1.png","contentUrl":"https:\/\/www.neuraldesigner.com\/wp-content\/uploads\/2023\/05\/logo-neural-1.png","width":1024,"height":223,"caption":"Neural Designer"},"image":{"@id":"https:\/\/www.neuraldesigner.com\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/x.com\/NeuralDesigner","https:\/\/es.linkedin.com\/showcase\/neuraldesigner\/"]}]}},"_links":{"self":[{"href":"https:\/\/www.neuraldesigner.com\/api\/wp\/v2\/blog\/3372","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.neuraldesigner.com\/api\/wp\/v2\/blog"}],"about":[{"href":"https:\/\/www.neuraldesigner.com\/api\/wp\/v2\/types\/blog"}],"author":[{"embeddable":true,"href":"https:\/\/www.neuraldesigner.com\/api\/wp\/v2\/users\/11"}],"version-history":[{"count":0,"href":"https:\/\/www.neuraldesigner.com\/api\/wp\/v2\/blog\/3372\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.neuraldesigner.com\/api\/wp\/v2\/media\/2634"}],"wp:attachment":[{"href":"https:\/\/www.neuraldesigner.com\/api\/wp\/v2\/media?parent=3372"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.neuraldesigner.com\/api\/wp\/v2\/categories?post=3372"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.neuraldesigner.com\/api\/wp\/v2\/tags?post=3372"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}