Skip to main content

'OneClick’ Manufacturing Services and Intelligent Machines

- By  Binil Starly, Atin Angrish and Paul Cohen
Edward P. Fitts Department of Industrial and Systems Engineering,
North Carolina State University, Raleigh, NC, USA


Abstract

The digitalization of manufacturing has created opportunities for consumers to customize products that fit their individualized needs which in turn would drive demand for manufacturing services. However, this ‘pull’ based manufacturing system production of extremely low quantity and limitless variety for products is expensive to implement. New emerging technology in design automation driven by data-driven computational design, manufacturing-as-a-service marketplaces and digitally enabled micro-factories holds promise towards democratization of innovation. In this paper, scientific, technology and infrastructure challenges are identified and if solved, the impact of these emerging technologies on product innovation and future factory organization is discussed. 
Computing technology has created orders of magnitude efficiency in the product life cycle but the skills required to design products have been largely confined to those skilled in the art and science of design and making of things. If barriers to lowering skills needed to engage in product design are reduced, increased expansion of the innovation ability of the consumer base will emerge. Products can be designed by anyone and not necessarily limited to those skilled in engineering and industrial design. Lowering the barriers would also mean that humans can focus on innate creativity, while computing tools work behind the scenes to improve designs and make recommendations while working collaboratively with the human. Once the product is designed and verified, consumers can potentially engage in a ‘one-click’ interaction to have the product assembly manufactured and delivered back to the client.

An example scenario is depicted in Figure. Other examples can include spare parts, personalized medical devices, customized consumer products, etc. Assistive design tools can enable a child to simply sketch out or verbally describe an imaginative toy while algorithms automatically detect design intent and generate a 3D Product model (a full Boundary representation - B-rep model) of the child’s version of the toy. Lowering the barriers would also allow a parent to engage in ‘one-click’ interaction to have the toy built. Algorithms would automatically determine the best capable and available manufacturers to make the product, including linking various manufacturers to suggest a price and delivery date. Machines negotiate and accept job order requests, fulfilling the order and initiating last-mile logistics to ship product to the client. With computing algorithms and physical machines performing the bulk of the operational steps, the focus of human activity will shift away from the requisite knowledge of design and manufacturing tools but towards innate human creativity in product design.



 Collaborative Design Bots for Product Design 


1) Product 3D Models and Natural Language descriptions
2) Joint Embedding of Text and 3D Product Models

3) Text-to-3D Model retrieval




Text-2-CAD paradigm (a) If datasets that leverage pairing of natural language descriptions of products are made available; (b) a joint embedding of textual description and associated 3D models can be made which clusters similar 3D shapes and text together through
semantic relationships, engineering specifications and/or shape descriptors; With a jointly embedded model, two application arise - (c1) Users can search for 3D product models based on detailed text descriptions of the product model. (c2) Design bots can be able to synthesize new designs based on textual descriptions provided by the user. 


 “One-Click” Manufacturing Service Marketplaces


While several manufacturing-as-a-service marketplaces have operated recently, these platforms are centralized intermediary brokers responsible for allocating orders to independent service providers. To realize the scenario of a “One-Click manufacturing” marketplace, several fundamental advancements must be made in the design of decentralized two-sided marketplaces that connects users with capable service providers. Manufacturing data remains the closely guarded asset that links the two sides, but its value can only be leveraged when such data is shared. Therefore, new scientific and technological advancements must be made in: 

1) Computing on encrypted data, such as 3D model data, machine information, pricing data etc., so that each party may share such information without fear of it being directly misused by either party; 
2) Incentivization mechanisms through microeconomic approaches and multi-class stochastic matching models to achieve fairness in the decentralized marketplace to ensure sustainable long-term operation; 
3) New techniques which move computation to the endpoint, rather than move data to a centralized system, such as through Federated Learning, would minimize issues surrounding privacy and security while still facilitating data-driven computational algorithms; 
4) Legal strategies and policy approach to ensure that digital manufacturing marketplaces can thrive in an interconnected world. Eventually “One-Click” Manufacturing intends to build a truly decentralized system that aims to reduce uncertainty in manufacturing networks by closing the gap between design users and the industrial base characterized by micro-factories, small job shops, and related manufacturing service providers.




Software-Defined Manufacturing Machines and Enhanced Portability


To distribute manufacturing capability through automated mini-factories and micro-factories, technological advances must be made in the design architecture of physical machines (3D
printers, CNC machines, laser cutters, robots and material handling systems). The separation of the software components and physical machine through a control layer is critical to make manufacturing resources digitally programmable and agile to allow dynamic adjustment of order traffic to meet changing needs through the development of Software Defined Manufacturing
(SDM) technologies

Conclusion


The impact of implementing tools and technologies for broad use among users, consumers and factory owners requires transdisciplinary collaboration across various domains of cyberinformation, design and manufacturing process automation, economics, business, law and public policy. It has the potential to impact rural economies across the world by directly connecting consumers with resource capable factories. The developed tools can enable new growth products and markets through mass entrepreneurship.  
Labels:

Comments

Post a Comment

Popular posts from this blog

DDSP: DIFFERENTIABLE DIGITAL SIGNAL PROCESSING

 -By Google Research {jesseengel,hanoih,gcj,adarob}@google.com Paper Link Code Demo Abstract Most generative models of audio directly generate samples in one of two domains: time or frequency. While sufficient to express any signal, these representations are inefficient, as they do not utilize existing knowledge of how sound is generated and perceived. A third approach (vocoders/synthesizers) successfully incorporates strong domain knowledge of signal processing and perception, but has been less actively researched due to limited expressivity and difficulty integrating with modern auto-differentiation-based machine learning methods. In this paper, we introduce the Differentiable Digital Signal Processing (DDSP) library, which enables direct integration of classic signal processing elements with deep learning methods. Focusing on audio synthesis, we achieve high-fidelity generation without the need for large autoregressive models or adversarial losses, demonstrating that DDSP enable...
Post a Comment
Read more

Capsule Neural Networks

-By Sara Sabour, Nicholas Frosst,  Geoffrey E. Hinton  Google Brain Toronto Paper Link Abstract   A capsule is a group of neurons whose activity vector represents the instantiation parameters of a specific type of entity such as an object or an object part. We use the length of the activity vector to represent the probability that the entity exists and its orientation to represent the instantiation parameters. Active capsules at one level make predictions, via transformation matrices, for the instantiation parameters of higher-level capsules. When multiple predictions agree, a higher level capsule becomes active. We show that a discrimininatively trained, multi-layer capsule system achieves state-of-the-art performance on MNIST and is considerably better than a convolutional net at recognizing highly overlapping digits. To achieve these results we use an iterative routing-by-agreement mechanism: A lower-level capsule prefers to send its output to higher...
1 comment
Read more

Detecting Anomaly in Big Data System Logs Using Convolutional Neural Network

By - Siyang Lu, Xiang Wei, Yandong Li, Liqiang Wang Department of Computer Science, University of Central Florida, Orlando, FL, USA School of Software Engineering, Beijing Jiaotong University, China Paper Link Previous week we did a scan of the survey of Anomaly detection   , that covers breath of the topic (Types of Anomalies, Types of Models and their applications). In this summary paper, we look in to the core Heavy lifting part of Detecting Anomaly in Big Data System Logs Using Convolutional Neural Network. Abstract Nowadays, big data systems are being widely adopted by many domains for offering effective data solutions, such as manufacturing, healthcare, education, and media. Big data systems produce tons of unstructured logs that contain buried valuable information. However, it is a daunting task to manually unearth the information and detect system anomalies. A few automatic methods have been developed, where the cutting-edge machine lear...
Post a Comment
Read more