As the growth engine of consumer markets shifts from the supply side to the demand side, manufacturing systems have grown far more complex than before. By deploying advanced sensing technology combined with AI algorithms, enterprises can boost data visibility and system controllability. Driven by the rising adoption of cyber-physical systems (CPS), the Industry 4.0 era of intelligent automation and smart manufacturing has arrived. This article analyzes the core development priorities for 2019, including collaborative robots, digital twins and predictive maintenance.

1. Collaborative Robots (Cobots) Gain Higher Application Value Driven by Sensor Advances

Traditional heavy-duty industrial robots once played a central role in factory production. However, amid fast-changing and volatile market demands, they have gradually been replaced by cobots due to high deployment barriers, substantial capital outlays, long return-on-investment cycles and rigid operational limitations.

Cobots enable enterprises to achieve simpler deployment and greater flexibility at lower costs. Most require no extra safety enclosures or extensive software suites, making them capable of adapting to ever-shifting manufacturing requirements.

Cobots are ideal for repetitive tasks performed alongside human workers that demand no manual craftsmanship, independent judgment or on-site contingency handling, while simultaneously enhancing workplace safety.

Cobots hold clear price advantages over conventional robots. Small and medium-sized enterprises (SMEs) form the primary growth engine of the cobot market. Industries requiring short design cycles and high product variability are major adopters, such as automotive and electronics manufacturing, which prioritize automated flexibility.

Industry Leader: Universal Robots Focuses on 3D Tasks and Global Expansion

Universal Robots retains nearly half of the global cobot market share, ranking first worldwide. The company launches new automation solutions to address manufacturing labor shortages, developing dedicated robot models and application platforms for so-called 3D jobs — Dirty, Dull and Dangerous work.

Beyond product R&D, Universal Robots aggressively expands into overseas markets, especially regions with low industrial robot density. Key target markets include China, the world’s largest robotics market; Malaysia, where manufacturing accounts for 98.5% of the economy yet Industry 4.0 adoption and robot penetration remain low; and India, which deploys merely 3 industrial robots per 10,000 workers.

Tech Advancements at Techman Robot: Deepened Core Machine Vision Capabilities

A Quanta subsidiary, Techman Robot launched its first commercial cobot at the end of 2016. By 2018, it surpassed Japan’s Fanuc (one of the four major industrial robot manufacturers) to become the world’s second-largest cobot vendor by market share.

Techman specializes in machine vision R&D. At Hannover Messe 2019, it unveiled the TM Operator series. The TM Palletizing Operator leverages visual inspection to automatically compensate for positional deviations of goods and pallets. Equipped with an intuitive graphical interface, it allows fast programming for users, delivering real-time work progress tracking and robotic arm status monitoring.

Techman further integrates hardware partners including Germany’s IDS (maker of Ensenso 3D cameras) and Switzerland’s Asyril (specializing in vibratory bowl feeding systems). Precise object identification enables deeper industrial use cases and expands its intelligent vision ecosystem.

Traditional Industrial Robot Giants Enter Cobot R&D and Production

ABB, a leading supplier of industrial automation and robotics, is a major cobot manufacturer. Its YuMi series includes single-arm and dual-arm collaborative robots, offering a lower-cost, more flexible alternative to ABB’s conventional industrial robots.

Powered by ABB’s Ability IoT suite, the platform interconnects robots and other equipment for precise monitoring and control. Data-driven analytics improve hardware performance, operational reliability and service life.

On peripheral hardware, ABB released wireless sensors for monitoring Dodge mounted bearings. The sensors track temperature and vibration to assess bearing health. Remote monitoring eliminates the need for maintenance staff to physically access hazardous bearing installation points for safety inspections. ABB continues to develop cobots on a modular design platform, enabling customized solutions and a wider range of cobot form factors and sizes.

Germany’s KUKA, another of the four top robotics conglomerates, focuses on Human Robot Augmentation (HRA). Cobots act as an extension of human operators to deliver superior precision and sensitivity. Powered by machine learning, these cobots interact with humans in real time and adapt to changing tasks without resetting, supporting fast, accurate and safe component assembly.

Intensified Industrial Competition Calls for Careful Product Development and Regional Layout

Driven by innovations in sensors, machine vision and AI integration, cobots are evolving toward enhanced safety and ease of use, taking over repetitive tasks once exclusive to human labor to generate greater industrial value.

Instead of fully replacing human workers, cobots form complementary partnerships with staff. By undertaking strenuous, repetitive work, they lift productivity and employee satisfaction, easing public concerns and resistance toward robot substitution. Manufacturers also equip engineers with automation skills and technical tools, even creating new job roles. Nevertheless, the influx of new suppliers has led to fierce competition across the sector.

2. Digital Twins Enable Real-Time Virtual-Physical Feedback with Diverse Cross-Industry Applications

A digital twin is a virtual replica of physical assets, processes and systems built via sensor data collection, widely recognized as a critical strategic technology.

Unlike conventional simulation tools, digital twins deliver real-time cyber-physical integration by establishing bidirectional links between physical and virtual models. Sensor data is transmitted, processed and analyzed instantly to generate feedback from the virtual replica, optimizing physical products and boosting commercial value.

The technology has been rolled out across multiple vertical sectors: healthcare uses it to manage emergency room waiting times and patient flow; construction leverages remote monitoring to cut operating costs; the energy sector applies it for real-time equipment oversight and power transmission & distribution regulation.

Within manufacturing, digital twins support product design, on-site production adjustment and new product development. They shorten development cycles to meet mass customization and low-volume production trends, extend component lifespans, enable rational production planning and precise process control, and optimize end-to-end business workflows.

Additionally, the installed base of wireless IIoT devices exceeded 20 million units in 2018 amid industrial automation transformation. Widespread connectivity and mature IIoT infrastructure enhance the accuracy of digital twins, laying the groundwork for smart manufacturing.

Digital twins deployed in smart manufacturing fall into three core categories:

1.Product digital twins: Allow manufacturers to conduct virtual adjustments, test and verify product functionality, safety and quality before physical production, drastically shortening overall development lead times.

2.Production digital twins: Focus on virtual commissioning to advance full digitalization and unmanned operation on production floors.

3.Operational digital twins: Collect operational data from products, machines and full production lines, using simulation to predict performance failures, energy consumption peaks and unplanned downtime risks.

Major digital twin solution providers include Siemens, Microsoft, GE and IBM. All platforms support virtual-physical integration to assist decision-makers, yet each vendor prioritizes distinct development directions:

• GE’s Predix asset and operation analytics center on direct equipment performance management.

• IBM’s digital twin products focus on full-lifecycle optimization.

• Microsoft Azure Digital Twins is primarily applied to factory and power grid facility management. Virtual modeling paired with software-defined hardware enables rapid deployment of IoT services via cloud platforms. Microsoft expanded the usability of digital twin technology by releasing IoT Plug and Play modeling language in 2019, with plans to integrate the upcoming Digital Twin Definition Language (DTDL) to enable plug-and-play monitoring and analytics within Azure Digital Twins.

• Siemens connects physical products, factories, machinery and systems via its MindSphere platform, centering its digital twin solutions on product design, on-site simulation and decision support. It also delivers solutions that integrate machine tool builders and operator workflow chains through digital twin technology. Tailored for CNC and additive manufacturing’s unique simulation flexibility requirements, Siemens launched the Sinumerik ONE dedicated digital twin system.

3. Predictive Maintenance Outperforms Reactive and Preventive Maintenance, Supported by Comprehensive Infrastructure

Predictive maintenance is a pivotal smart manufacturing technology. It optimizes maintenance schedules by analyzing production data and real-time equipment status monitoring, effectively preventing unplanned downtime, cutting maintenance costs and maximizing asset uptime to boost output.

Modern factories operate around the clock, pushing downtime costs steadily higher. A single hour of production halt may cost manufacturers USD 100,000 to 300,000, while unplanned outages can trigger losses reaching millions of dollars. By optimizing maintenance schedules, minimizing unexpected downtime, extending equipment service life and lifting staff productivity, predictive maintenance has become a flagship Industry 4.0 application.

Predictive maintenance relies entirely on IIoT infrastructure, consisting of five core components: data acquisition sensors installed on equipment; communication systems transmitting data between monitored assets and central data hubs; centralized data repositories storing and processing cross OT/IT system data; predictive analytics algorithms; and data visualization interfaces for maintenance and process engineers to diagnose equipment conditions.

Enterprises including Siemens, SAP and GE have deployed predictive maintenance at scale in smart manufacturing. SAP’s solution merges sensor data with ERP and Enterprise Asset Management (EAM) business information on an IIoT foundation. Anomaly detection, spectrum analysis and machine learning algorithms optimize asset upkeep. The platform integrates with SAP S/4HANA and third-party maintenance execution systems to expand service coverage, lower maintenance expenses, improve asset availability and visualize predictive analytics results.

In 2019, IBM launched its Maximo Asset Performance Management suite. Its Predictive Maintenance Insights module applies statistical models and machine learning to forecast asset health metrics, including failure timelines, failure probability, root cause drivers, degradation curves and anomaly detection.

IBM also extends predictive maintenance to smart city scenarios. It partnered with Atlanta’s transportation authority to shift asset management from passive tracking to proactive failure prevention. Most recently, IBM announced collaboration with Denmark’s state-owned infrastructure operator Sund & Bælt (S&B). The newly released Maximo for Civil Infrastructure integrates structural sensor data, worker wearable device readings, drone-collected information and meteorological data to extend the service life of aging bridges, tunnels and railways.