Intro to Business › Section J · Operations management › Topic J.6
Technology on the factory floor
Computers now design products, drive machines, inspect output and predict when equipment will fail. This topic explains the main technologies that automate production, how firms judge whether they are worth buying, and how the same ideas reach banks, shops and hospitals.
Why firms automate
Automation replaces or assists human effort with machines controlled by computers. Firms automate to raise output, cut cost per unit, improve consistency and remove people from dangerous, dirty or highly repetitive work. Automation also changes the kind of work people do: fewer jobs moving parts by hand, more jobs programming, maintaining and supervising machines. This topic covers six families of manufacturing technology: computer-aided design and manufacturing, additive manufacturing, robots, flexible systems, computer-integrated manufacturing, and sensors with artificial intelligence. Each is expensive to install, so firms compare the investment with the savings it brings, as the worked examples in this topic show.
CAD, CAM and 3D printing
In computer-aided design (CAD), engineers draw products on screen in three dimensions, view them from any angle and test them by simulation before a physical part exists. Computer-aided manufacturing (CAM) uses computers to plan production and send instructions directly to machines, such as computer numerically controlled (CNC) lathes and cutters. CAD/CAM links the two, so a design change flows straight into machine instructions. Additive manufacturing, or 3D printing, builds objects layer by layer from a digital file, an approach Chuck Hull patented in the 1980s with stereolithography. It is used for prototypes, tooling, custom medical parts and spare parts printed near where they are needed.
Robots and cobots
A robot is a computer-controlled machine that can be programmed to carry out tasks on its own. The first industrial robot, the Unimate, began work at a General Motors plant in New Jersey in 1961, lifting and stacking hot die-cast parts. Fixed robots, usually jointed arms, weld, paint, assemble and pack; mobile robots move materials around warehouses. Robots suit tasks needing speed, strength, precision or endurance. Traditional industrial robots work behind safety fences. Collaborative robots, or cobots, are designed to work safely beside people, sensing contact and slowing or stopping; a cobot might hold a part steady while a worker fits it, or handle the repetitive half of a task.
Flexible and integrated manufacturing
A flexible manufacturing system (FMS) links computer-controlled machines and automated material handling so that the same equipment can switch quickly between different products. It is expensive to install, but once running it needs little labour, gives consistent quality and can be reprogrammed cheaply, which suits small batches of varied products and mass customisation. Computer-integrated manufacturing (CIM) goes further, connecting production machines with the computer systems for design, inventory, scheduling and purchasing. With CIM, a part redesigned in CAD is sent automatically to the machines that make it and to every department that must plan for the change. It is the factory-floor counterpart of ERP, covered in the topic on resource planning.
Sensors, AI and digital twins
Modern machines carry sensors that report temperature, vibration, speed and output, linked through what is often called the industrial Internet of Things. Artificial intelligence analyses this data in several ways. Predictive maintenance spots patterns that come before a breakdown, so a part is replaced just before it fails rather than on a fixed timetable or after it breaks. Machine vision inspects products with cameras and flags defects faster and more consistently than a tired human eye. A digital twin is a virtual model of a machine, line or whole plant, fed with live data, on which managers can test a change, such as a new schedule, before trying it for real. These tools depend on reliable data and on staff who can act on what they show.
Measuring equipment effectiveness
A common way to judge how well equipment is used is overall equipment effectiveness (OEE):
Availability is the share of planned time the machine actually runs. Performance compares actual output with what the machine could make at its ideal speed during that running time. Quality is the share of output that is good. Because the three multiply, modest losses in each add up: 90% in all three gives an OEE of about 73%. OEE shows where to aim technology. Frequent breakdowns point to predictive maintenance, slow cycles to better programming or tooling, and defects to machine vision or process control.
Worked example: OEE on a packing machine
A packing machine is planned to run 400 minutes in a shift. Breakdowns and changeovers stop it for 40 minutes. At its ideal rate of 1 pack per minute it could make 360 packs in the remaining time, but it makes 320, of which 304 pass inspection. 1. Availability: \(360 \div 400 = 0.90\). 2. Performance: \(320 \div 360 \approx 0.889\). 3. Quality: \(304 \div 320 = 0.95\). 4. OEE: \(0.90 \times 0.889 \times 0.95 \approx 0.76\), or 76%. 5. Check another way: good packs at ideal speed, \(304 \times 1\) minute, divided by planned time, 400 minutes, also gives 0.76. 6. The largest loss is performance, about 11%, so speed is the first problem to investigate.
Worked example: is a robot worth it?
A furniture maker considers a robotic sanding cell costing 240,000 dollars installed. It would save 90,000 dollars a year in labour and rework but add 15,000 dollars a year in maintenance and power. 1. Net annual saving: \(90{,}000 - 15{,}000 = 75{,}000\) dollars. 2. Payback period: \(240{,}000 \div 75{,}000 = 3.2\) years. 3. If the firm's rule is to accept projects that pay back within 4 years, the cell qualifies. 4. Limits of the measure: payback ignores savings after year 3.2 and the timing of cash, so firms also use methods covered in Section P's topic How organizations use funds. 5. Check: forgetting the extra running costs would give a misleadingly quick \(240{,}000 \div 90{,}000 \approx 2.7\) years.
Automation beyond the factory
Service firms automate too. Banks use cash machines, the first of which was installed by Barclays in north London in 1967, along with mobile apps and online banking. Retailers use point-of-sale terminals that scan barcodes, record each sale, update inventory and trigger reorders; the first supermarket scan of a Universal Product Code took place in Ohio in 1974. Warehouses use automated storage and mobile robots, and hospitals use robots to sort samples and assist some surgical procedures. As in factories, automation in services raises questions about jobs, privacy and what happens when systems fail, and it works best when it frees people for tasks that need judgement and care.
Common mistakes to avoid
First, confusing CAD, which designs products, with CAM, which controls the machines that make them. Second, describing a cobot as any robot; cobots are designed to work safely beside people. Third, adding the three OEE factors instead of multiplying them. Fourth, computing payback from gross savings and ignoring the new running costs. Fifth, treating payback as a full measure of value; it ignores later savings and the timing of cash. Sixth, assuming automation simply removes jobs; it also creates and changes them, which is why firms invest in retraining.
Key takeaways
CAD designs and tests products; CAM controls machines; CAD/CAM links them; 3D printing builds parts layer by layer. Robots, starting with the Unimate in 1961, suit fast, strong or precise tasks; cobots work beside people. FMS switches equipment between products; CIM links production with design, inventory and purchasing. Sensors and AI enable predictive maintenance, machine vision and digital twins. OEE = availability × performance × quality. Payback = investment ÷ net annual saving. Services automate through cash machines, point-of-sale systems and robots.
Flashcards
Flashcard
Computer-aided design: drawing and testing products on screen
Test your understanding
Question 1 of 5
A machine has availability 80%, performance 90% and quality 95%. What is its overall equipment effectiveness?
Question 2 of 5
A robot costs 180,000 dollars. It saves 70,000 dollars a year in labour but adds 10,000 dollars a year in running costs. What is its payback period?
Question 3 of 5
A plant uses vibration sensors and AI to replace a bearing shortly before the data suggest it will fail. What is this called?
Question 4 of 5
Which description fits a collaborative robot, or cobot?
Question 5 of 5
What does computer-integrated manufacturing (CIM) add to a flexible manufacturing system?
