envisioned for NDE in the new industrial area of Industry 4.0. These concepts have inspired forward-looking discus- sions, and some significant activities have now been initiated to move these concepts forward. This paper is intended to contribute further to these discussions and to present the developing ideas of the authors. In its simplest form, NDT is almost as old as mankind. Testing could involve listening to the cracking of a bent tree branch or the response to knocking on a nutshell to determine if it is hollow this eventually evolved into a simple acoustic emission testing inspection that involved tapping a pot or a visual examination to assess aesthetic quality. Testing became a way to check on the quality of workmanship during the 19th century and began to evolve from about 1870 into what is now called NDE. At the beginning of industrialization, steam power replaced human and animal muscle power for many tasks, and products were simply inspected randomly by using human senses, but such testing was found to be inadequate when new technologies were produced and items such as boilers failed. The basic idea for providing a nondestructive test for quality has not changed a lot since the first indus- trial revolution and up until the last decade of the 20th century however, the instrumentation we use and the way we apply the techniques now depend upon more complex technology and industrialization. In recent years, the basic idea has evolved from inspecting to minimize the occurrence of discontinuities, to life concepts such as damage tolerance. With more advanced equipment, we can now look for smaller and hidden anomalies and plan inspection intervals based on estimated damage growth rates so as to assure safety. As advancing technologies seek to push materials to their performance limits, NDE becomes a risk-management tool. Fundamentally, this philosophical approach of risk management becomes a more active and real-time function with NDE 4.0. Advances in science have provided the NDT community with a diverse range of new tools and capabilities. Looking through transparent or opaque objects was possible before the discovery of X-rays. However, with the availability of X- and gamma rays, we extended the applicable wavelength range of elec- tromagnetic radiation beyond what was available by using light. This advancement allowed the user to look through materials that were not optically transparent. The development of sonar and then ultrasonic NDT extended the range of frequencies that were available beyond human hearing and to a range above 20 kHz. This involves shorter wavelengths and therefore allows better local resolution. Such ultrasonic testing (UT) is now seen as just part of an acoustic spectrum that includes applications in surface acoustic wave devices for electronics and exploration seismology. NDT now also utilizes technologies that employ heat (thermal inspections), light and electromagnetic waves (visual, liquid penetrant, magnetic particle, X-ray, and eddy current techniques), and sound (acoustic emission and ultrasonic) in all their various forms (Ahmad and Bond 2018). Looking forward, we can now talk about NDE 4.0, which does not mean that we invent a new NDE technique, but rather that we make the capabilities of NDE available in new implementations to meet upcoming challenges and to prepare for the mid-21st century. NDE and Progress in Industry Major innovations in technology can be seen in Figure 1, which provides a simple timeline for the history of technology. In the 19th and into the early 20th century J U L Y 2 0 2 0 • M A T E R I A L S E V A L U A T I O N 795 Age of (artificial) intelligence Smart cross-linked systems replace human decisions Age of information Electronic machines replace human memory Age of mechanization Mechanic machines replace human muscle power 2000 1900 1800 1700 around 2010 Internet of Things around 2000 Complex smart sensor systems 2000 Sensor networks Local and global networks 1960 Sensors and amplifiers Simple sensors 1909 Nobel prize for wireless telegraph 1901 Picture telegraph 1833 Electromagnetic telegraph (Morse) 1792 Wing telegraph 1769 James Watt 1700 Thomas Newcom Heron of Alexandria 1st century Figure 1. An attempt to characterize the 19th, 20th, and 21st centuries by showing typical innovations in technology. This does not completely correlate to the recognized industrial revolutions (see Table 1). It illustrates that important innovations have often been made many years ahead (some of them cannot be exactly dated). It is the authors’ opinion that smart machines that have the ability to learn, solve complex tasks, and make decisions on their own or work with humans “hand in hand” will be the key technology of the future.

796 M A T E R I A L S E V A L U A T I O N • J U L Y 2 0 2 0 (1800–1950s), machines replaced muscle power (humans or animals), initially using the steam engine and then later electric and gasoline engines. These changes enabled a significant increase in productivity, and this period of history is well known as the Industrial Revolution. We can characterize this time period as the “age of mechanization” and call it Industry 1.0. During this time, NDT was characterized by techniques that used human senses (visual/sound) and then simple techniques to enhance the signals that can be detected by humans (including liquid penetrant testing and magnetic particle testing). The 20th century is considered to be the “age of information.” The Nobel Prize in Physics 1909 was awarded jointly to Guglielmo Marconi and Karl Ferdinand Braun in recognition of their contributions to the development of wireless telegraphy (Nobel Foundation 1967). It subsequently took 100 years for this first step to evolve into today’s cellphones and global telecommunication. However, the fundamental capabilities needed to enable this age were electricity and later electronic machines (computers), which became tools used to supplement the capabilities of the human brain. In manufacturing, these develop- ments began with the application of electricity to what created the first physical network and the introduction of conveyer techniques, which established mass production (Industry 2.0). In the second half of the last century, mass production was improved further by the introduction of electronic-controlled and automated production (Industry 3.0). This prompted the need for reliable NDE to provide 100% inspection for large quantities of parts and the quantification of NDE performance using probability of detection (POD). With the development of electronics and portable computers, it was also possible to develop automated inspection techniques and increasingly replace analog instruments with digital ones. This was essential in enabling various advancements in UT, such as phased array UT (PAUT), that we use today. There were corre- sponding advances in most NDE technologies. Both X-ray and electromagnetic testing techniques where pushed forward by developing new sensors and data analysis procedures, including image processing and computed tomography (CT) (Thompson and Chimenti 1980–2019). What Comes Next? It is the authors’ opinion that the next 50 to 100 years will be characterized by increasing capabilities to collect and manage digitized data, which are produced by various forms of big-data processing and ME FEATURE w nde 4.0: challenges and opportunities TABLE 1 Impact of industrial revolutions on NDE* Industrial revolution NDT/NDE NDE techniques introduced 1st (1750–1850) 1.0 • Visual tesing • Mechanization • Using human senses for • Acoustic emission techniques • Replacement of muscle power random inspection • Unique components 2nd (1850 –1960) 2.0 • Liquid penetrant testing • Mass production • Enhancing detectability of human senses • Magnetic particle testing • Assembly lines (for instance, for surface-breaking cracks) • Electrical energy • 100% manual inspection of selected safety • Identical components relevant parts 3rd (1960–today) 3.0 • Radiographic tesing • Automation • Using physical effects, radiation, or fields • Ultrasonic testing • Electronic control and data processing to detect discontinuities, measure material • Electromagnetic testing • Multifunctional microelectronic systems properties • Manual or automated inspection • 100% inspection of large quantities of parts 4th (next) 4.0 • Computed tomography • Cyber-physical systems • Use of cyber-physical systems • Phased array ultrasonic testing • Learning and decision-making machines (cloud computing, Internet of Things, • Optics • Individual custom-tailored components modeling) • Thermal/infrared testing • Continuous monitoring of manufacturing • Terahertz processes or components in service • Large-volume data files (3D images) *adapted from Meyendorf (2017)