A White-Paper Advocating an Industrial Robotics Network for Low-End Manufacturing
This paper was written in 2012 with ideas formulated in 2009 and 2010 and revised in 2017. Methods and systems supporting the architectural concepts presented below are supported by patent rights in the United States, the United Kingdom, Australia and internationally. This white paper is a precursor to method and systems for an “Autonomous Supply Chain Model.” The autonomous supply chain model has both civilian application as well as application in defence. See also Methods and Systems for an Autonomous Supply Chain and supporting patent Methods and Systems for a Robotic Capability Model.
The Internet has done globally for information what the printed circuit board has done for electronics and electronic components: establish a universal standard of interchange and composition and thus provide a mechanism for scalability. Australia has recently capitalized on this global trend by starting an ambitious project to field a National Broadband Network that will offer 1 Gigabyte speeds to 93% of Australian households and businesses alike. “Down Under” aims to become the world’s smart country through a significant infrastructure upgrade. Yet the Internet of today only facilitates the processing of information & the exchange of ideas.
While these are powerful concepts, the Internet of today cannot be used to make and download manufactured goods. This white-paper proposes a strategy to achieve just that.
It also argues that industrial democracies are engaged in an epic struggle because the partnership between democracy and industry of today has broken down as the result of Information Technology. It then proposes leveraging the same technology in a seemingly counterintuitive move to restore this partnership.
The key points of this essay are summarized below.
1) Democracy arose in response to the industrial revolution giving economic clout to an educated middle class rather than, as is commonly believed, democracy imparting high living standards.
2) Information Technology removed the economic clout of the middle class on a global scale, thus altering economics in favour of population rich systems without a need for democracy.
3) Progress that human civilization has made in favour of the principles of balance-of-power and government-for-and-by-the-people is at risk.
4) A Solution is proposed through accelerating information technology development to once more favour democratic systems – a pathway to a smart, post industrial economy.
Democracy arose in response to the industrial revolution giving economic clout to an empowered middle class rather than, as is commonly believed, democracy having empowered the middle class
Democracy was never a guarantee for good government, just a fair way of selecting one. Nonetheless, democracy must stand as the most significant advancement human civilization has made in modern history.
The dawn of democracy was to predate its dominant era by a significant amount of time. The Greeks spearheaded democracy in the year 508 BC in what then comprised the Athenian city state of Attica. Profoundly non-democratic systems were to follow, embodied by Roman emperors claiming divine status, and the autocratic emperors of Germany, Japan and China. Indeed, the mega-political age of the monarchy was to last until the 18th & 19th century when liberal democracy was to begin sweeping much of the Americas and Europe. This period coincides with the Industrial Revolution from the 18th to the 19th century where major changes in agriculture, manufacturing, mining, transport, and technology had a profound effect on socioeconomic and cultural conditions, a process that started in the United Kingdom, then spread throughout Europe, North America, and eventually the modern world. The Industrial Revolution was to embody the alignment of circumstance that propelled democracy onto centre stage.
Consider the steelworkers of Germany, Japan & the United States in the 20th century: These steelworkers typified the common man of the industrial age. For example, a typical operator of a lathe, a steel-machining tool, was a highly educated worker. To operate a lathe machining parts for assembly in a steel factory, the operator of a lathe had to be a) literate, b) numerate, c) proficient in geometry, & d) knowledgeable in metallurgy and elementary physics. The education of this individual would have involved primary school, secondary school and a two to three year apprenticeship. Starting around age six and concluding around age 16-18, this schooling would have comprised about 10-12 years of education, a significant economic investment. With the bulk of the workforce of an industrial nation involved in like tasks with like requirements, an educated middle class emerged, a middle class that possessed economic clout and economic clout always compels political representation. In this age, only systems that facilitated political representation of the middle class would prosper, because only such systems provided the catalysts that aligned economic equations with power equations. In a process of natural selection among nations, democracies were propelled to wealth. It was this basic alignment of the equations of power and economics that allowed democracy to emerge as the dominant force on the globe.
Information technology removed the economic clout of the middle class on a global scale, thus altering global economics in favour of population rich systems without a need for democracy.
An invention marked the end of the dominance of the middle class: The computer. Driven by Word War II ballistics calculations, an invention entered modern economics that would erode the economic clout of the middle class and usher in an era of political frustration, a feeling that would rival the frustration felt by medieval peasants at the obvious corruption of the clergy in the feudal age. A feeling would prevail that the political leaders no longer represented the people. Simply, information technology and the computer made it possible for the lathe operator of the industrial age steelworks, to be replaced by a pair of nimble hands – without education, without rights, without skills, without representation. These nimble hands need not understand the metallurgy or geometry underlying the operation of the lathe. This more than anything is at the root of the world trade imbalance that now has come to threaten the global economy and explains why so little political representation and social maturity follows economic success in the developing world. The hundreds of millions of nimble hands operating the assembly lines in sweatshops of the developing world are expendable and cannot hope for more political power and freedom than the peasant of the feudal age. The factories at which Apple IPods are made, for example, have exhibited so many suicides and suicide attempts; management reportedly solicited the services of a monk to try to expel the bad energy. The age of the new serf without rights and representation is upon us – truly a great leap backward for mankind.
The developing world of today is at the beneficial end of an altered equation of economics that places countries with millions of nimble, expendable hands at an advantage over the developed world, because fundamentally, the bulk of the educated middle class is no longer needed in the post industrial economy. The middle classes’ continued demand for political representation and the costs and inefficiencies that this representation incurs in the developed world places democratic systems at an inherent disadvantage over more autocratic systems. Thus the partnership that existed between industry and middle class is broken. Consequently there is an outflow of capital from developed world to the developing world in which the middle class enjoys few rights. Otherwise free capital and free corporations willingly bond themselves to unfree systems, because simply put, the price is right.
The gains human civilization has made in favour of the principles of balance-of-power and government-for-and-by-the-people are at risk
In the year 1215 the Magna Carta required King John of England to proclaim certain liberties, and accept that his will was not arbitrary, for example by explicitly accepting that no “freeman” could be punished except through the law of the land. Though repealed in part and revised over the ages, this charter enshrined a principle in English law that was to lay the foundation for the stability of the British Empire in the centuries to come. This principle recognized that power must be confined in a system of checks and balances. Power on its own will corrupt those who exercise it with arbitrary governance. In the word of the vernacular expression: Power corrupts; absolute power corrupts absolutely. Over the centuries to follow, Europe would convulse in an endless series of bloodletting that would climax in two world wars. Hundreds of millions of people have died to arrive at the order Western Europe largely takes for granted today: the longest period of peace in modern history where power is derived by popular choice – for and by the people. One of Western Europe’s last dictators, Fransico Pauline Franco of Spain died in the year 1975. Josip Broz Tito of Yugoslavia died in the year 1980. Today, Slovenia, a former Yugoslav republic is a full democratic member of the European Union.
The majority of the world now holds this truth to be self-evident, that only government elected by popular democratic choice may call itself legitimate.
Yet the bloodletting and the ground gained by human civilization over the course of 800 years is fundamentally at risk if the world gets the economics and politics of the next 10 to 20 years wrong and regresses to systems of autocratic rule. For this reason humanity must re-align the synergy between commerce and democracy.
Restoring the balance to favour democratic systems
High-end manufacturing has remained the main stay of modern democracies. Examples of this stronghold include aircraft, automobiles, defence industry and others. In these industries, nimble hands are not enough. These industries still require skilled labour, the kind that formed the educated middle class of developed nations. The United States for example, which imports almost all low end goods from toys to ballpoint pens, is still the leader of high end manufactured goods – together with Germany and Japan. The key point is that high-end manufacture transcends the domain of mere assembly and packaging. It requires extensive domain knowledge and a literate, skilled work force. Low-end manufacturing, by contrast, is characterized by a focus on assembly. The question is whether workers in the developed world once more wish to be employed in this kind of task?
Consider that low-end manufacturing has been forever altered by information technology. If we repatriate these industries, we cannot expect more than what is lovingly referred to as “Mc Jobs,” the counterpart to dead-end service industry work offered by fast food restaurants. As mentioned previously, low-end manufacturing has been de-skilled. Therefore low end manufacturing jobs would expect to pay significantly less than their pre-outsourcing counterparts. Nonetheless, we will want to repatriate low-end manufacturing such that we a) maintain a large scale manufacturing power base enabling military strength and b) so as to reduce unsustainable trade deficits built on the false premises of free trade. To elaborate, the benefits of free trade can only be materialized when comparative advantage, or specialization, is leveraged to increase the nation’s bottom line, i.e. a trade surplus. Free trade resulting in a trade deficit only benefits entities that span both the beneficiary nation carrying a trade surplus and the loser nation carrying the trade deficit. No sound, common sense, economic argument can be put forward, from a national perspective, for a sustained trade deficit – under the guise of free trade or otherwise. Current levels of trade deficit jeopardize the stability of the world economy. Another fallacy may be found in the proposition that service industries such as financial services or intellectual property based industries, as may be found in software copyright or patents to designs, can effectively replace manufacturing. Emerging economies do not benefit from enforcing legislation surrounding such notions, as doing so would add liabilities to their international balance of trade. Therefore, while emerging economies sell us manufactured goods, our ability to sell service based “products” to emerging economies is inherently limited, thus further contributing to a negative balance of trade.
Therefore, if we are to revert trade imbalances, we must reshore low-end manufacturing and do so in a manner that makes economic common sense while re-empowering the middle class. To this end, and this is the key proposal of this essay, it us suggested that we, in what may appear like a counterintuitive move, accelerate information technology to embrace an industry of networked capabilities – via robotics. Robotics lies at the high-end of manufacturing. It also extensively leverages information technology. Both areas are still the mainstay of democratic systems.
Why is it that a child’s toy or a plastic pen cannot be cheaply assembled by robots? If Google can drive a car across California that is robotically controlled, then surely a child’s toy can be assembled robotically. If done domestically, such manufacture would save on transport costs and therefore greenhouse emissions that otherwise contribute to global warming — it would benefit the environment as a whole, save import duties and eliminate other overheads associated with outsourcing, such as slower time to market, etc. Finally, it would expand the high-end manufacturing sector in production of assembly robots as well as create information technology jobs, jobs to build infrastructure, etc.
What is proposed is an autonomous manufacturing network that can be used to model various simple manufacturing processes. Advantage would go to the most skilled and to the most educated economies in the world: precisely the kind of economy produced by an advanced democracy.
A suitable strategy might be to target, through subsidies or tax advantages, individual industries in a phased approach. One might start with sports shoes, for example, currently a mainstay of Vietnam: develop the robots and information technology required to automatically manufacture sports shoes. After sports shoes, expand to textiles, then toys, then in an iterative fashion expand networked manufacture to all other low-end industries. Broadly speaking, today’s Internet caters for information only. The proposed national manufacturing network would allow assembly of physical goods. Key advantages offered by a robotic national manufacturing network might include the following:
1) Consistent Quality – The quality of goods on the shelves of our supermarkets has become abhorrent. Factors behind this are that “hands” make mistakes. A lot of “hands” make a lot of mistakes. Low-end manufacturing has been inherently deskilled. Unskilled hands make even more mistakes. A lot of unskilled hands pose an unmanageable quality control problem. Humans in systems of centrally planned economies that must report growth above all else make yet more mistakes. A machine operated manufacturing sector would solve this problem and warrant consistent quality levels.
2) Sub Specialization leaning on the lessons gleaned from the Chinese manufacturing sector. The Chinese manufacturing company operates differently from Western companies. Typical Western companies have mastered production, supply and marketing. Western brand names tend to control both manufacture and wholesale, if not retail of their products. Consider Apple, the company selling IPhones and IPods as well as notebook and desktop computers. Apple controls their products to the retail outlet in popular streets of major cities. This means that the company is able to capitalize on profits of every stage of the product life cycle. Chinese companies, by contrast, tend to manufacture “on-behalf-of” other companies, who then realize the real profit when the exported goods are taken to the market abroad. The Chinese manufacturer realizes but a tiny sliver of the eventual retail price of the product. Nonetheless, taking a tiny sliver of the retail price of manufactured goods the world over, is a substantial amount – about 8% of world GDP by all reckoning. The “on-behalf-of” model has clearly worked for China. What is advocated here is a similar specialization but on a much finer granularity. One might envisage one company offering cutting services, another offering assembly services and a third offering packaging services, all connected by the proposed industrial network. The product vendor enlists the services of all three offerings, routing their materials and products along the network from manufacturing stage to manufacturing stage, paying for both services used based on volume and complexity as well as for transport of their products along the network. The more optimal their process, the cheaper the end product will be.
3) Impossible to outsource non-domestically – An industrial network of machines and robots is a physical structure, not unlike the interstate highway system. As such it would be an ideal tool for policy makers the world over who seek to inject stimulus packages into their economies and to maximize local returns on local tax Dollars, Pounds or Euros.
Few large-scale achievements on a national scale are attributable to corporate initiative. This may be explained simply in that corporations exist to generate profits for their shareholders. Consider the national interstate highway system, modelled on the German Autobahn. Financed by the Federal government of the United States, it laid the foundation for what would become a way of live for Americans: the automobile. While the automobile is perceived as a very private and individualistic icon, the infrastructure to support the automobile was created through public initiative. No privately built roads – or toll roads – created a significant road network across the United States or any other developed country. Similarly, one of the greatest economic success stories of the last century, the Internet, was funded through public initiative: the ARPANET. Once more did private enterprise benefit from infrastructure created through public funds. The strategy proposed for a low-end manufacture robotics infrastructure follows this pattern: create a national network of robotics through public funds that private enterprise can lease for the assembly of products.
Some aspects of the architecture of the proposed national manufacturing network are described below:
1) Modular – A modular interface allows connecting robotic elements with different purpose, e.g. connect assembly robots to packaging robots. Any number of specialized robots is conceivable, dye and paint robots, sewing robots, etc. The strength of the proposed modular architecture lies in the arbitrary combination of these robots and the compositionality of their output – see Functional Paradigm below.
2) Transport Network – Work units may be routed dynamically between robotic workstations. This would enable a process to define such simple routes as assembly, painting followed by packaging. A suitable notation to drive this process might be the BPMN (Business Process Modelling Notation). Routing of work units could be done locally within the same building or shop floor or even across town or in between cities. One conceivable mechanism would be a monorail transporting work units from suburb to another. Monorails are cheaper to build than subways, may be integrated with existing road networks and could be reasonably small scale given that no human passengers need to be transported.
3) Session Orientation – work sessions define what robots do for a batch of work items. For each session, specific software is loaded into robots that perform functionality from within a specific domain, e.g. assembly or packaging, but whose hardware is otherwise product agnostic. Each session therefore defines product specific behaviour in software. As manufacturing is process or function oriented, functional programming lends itself particularly well to this domain.
4) Compositionality @ Hardware Level – the modular architecture of the network should allow processing of container structures in the place of work items, that is parent level robots should be able to process aggregates consisting of a work unit accompanied by another smaller robotic unit, that can be invoked by the parent level robotic unit. This facilitates the equivalent of what is known as dependency injection or strategy design pattern in software engineering and provides a mechanism functional scalability.
5) Layered Service Oriented Architecture – the architecture of the network would probably want to follow a layered model similar to that employed by computer networks and the OSI Open Systems Interconnection) model. Such a layered architecture would allow for separation of concerns such as transport, robotic domains, product specific session handling and others. This layered architecture also would need to cater for separation between consumables and non-consumables within the network. A cutting robot, for instance, is liable to wear out blades. The management of such consumables and charging for their use needs to be part of the architecture.
6) Functional Paradigm – A relatively recent paradigm in software development may lend itself here more than mainstream software development methods: functional programming. Unlike Object Oriented software development, functional programming models systems based on high-level functions within a system and stresses their compositionality that can be reasoned about at a mathematical level. As such, large-scale systems can be reasoned about and remain tractable, precisely the kind of behaviour one would want to see in a national infrastructure.
7) Lateral Scalability via Parallelism – task and work unit parallelism and the synchronization thereof should be a functional design goal that will permit lateral scalability of the network
8) Job Creation Across a Wide Spectrum – whereas information technology, or IT, has traditionally led to job creation only at the high end of the spectrum, a national manufacturing network would create jobs across the full social spectrum, ranging from construction to tool making, to information technology, robotics, as well as service and maintenance of both simple and advanced systems. This notion is something that ought to appeal to communities from the Rust Belt of the United States to the Ruhr Valley of Germany and the English Midlands.
This proposed strategy has the potential to turn even smaller countries into pillars of manufacturing and leverage global trade to effect significant trade surpluses, instead of deficits. Sceptics might argue that manufacturing and assembly of goods is too broad a vision and the field too diverse to benefit from significant automation in a cost effective manner. But consider that the same argument might have been advanced for the Internet and information processing. Information is a general concept and the kinds information are diverse and varied. Yet the Internet successfully handles everything from personal mail, social networking to electronic commerce, banking to numerous forms of entertainment. A carefully designed, layered compositional architecture can achieve the same for manufacturing that the Internet has achieved for information.
The potential for the manufacturing sector is even more far reaching when combining an information network like the Internet with a manufacturing network. Consider customer-designed products that are configurable via a web interface and are then routed as orders through a network of robotic assembly and processing stages. For example, rather than going to a department store’s outdoor section and purchasing an off-the-shelf bicycle, a customer might design their own bicycle by choosing compatible frame, seat, wheels, colour, etc. to suit the customer’s budget and personal wishes. The customer might then submit the order and the manufacturing processes are initiated with parts and assembly distributed across vendors and process providers. Every aspect, from component routing to assembly and packaging is automated and the final product is delivered to a local outlet near the customer’s residence. Such outlet might take the form of what is known as Sam’s Club in the United States or was known as a chain called Index in the United Kingdom. Such stores provide a warehouse retail model where customers order products, which are delivered without the benefit of a retail floor. Advantages are lower real estate costs and less shoplifting. The paradigm described represents the closest equivalent to downloading physical goods from a network rather than merely information.
Methods and systems supporting the architectural concepts presented above are supported by patent rights in the United States, the United Kingdom, Australia and internationally. Detailed solutions, simulations and prototypes will be presented in articles to follow.
Sydney, Australia, 2012