ContentsAbstractIntroduction1. Static Application upgrade for sustained yield stability2. Data Analytics/Mining/Visualization3. Array placement upgrade to reduce losses4. Acuity system utilised for array detection5. Augmented reality as a training tool6. Tween pad upgrade for increased yield and throughputConclusionReferencesBibliographyAbstractIndustry has evolved rather quickly over the past century and a half. Two world wars pushed technology so far that we went from the Wright Brothers first controlled, sustained flight of a powered, heavier-than-air aircraft in 1903, to landing a man on the moon in 1969. Such technological breakthroughs increased the necessity for more advanced production systems. As the world became increasingly smaller through globalization, they need to be more efficient and competitive in what is now a truly global market. Technology is constantly evolving as are the techniques required to apply these advancements to the production systems. Manufacturing has never been more efficient and as we move further on into the future it will never be as inefficient as it is now.IntroductionDuring the latter half of the 19th century, there was a rapid growth of industrialization in the United States, Britain, and Germany. Advancements in steel manufacturing technology and manufacturing practices meant steel could be manufactured at a competitive rate. This gave way to the extensive building of railroads that could transport mass quantities of material at relatively cheap cost. This extensive railroad network also saw the development of larger urbanized cities and the migration of people and ideas. During this time there were technological advancements in the mechanization of manufacturing practices and the breakthrough in electrification.  Such advancements lead to mass production, assembly lines and then the study of industrial practices to determine the most efficient and cost-effective way to remain competitive in the rapidly growing markets.To remain competitive in the global environment, it is important for an organisation to continuously look for ways to improve efficiency and productivity. (Groover, 2014) (International Labor Organisation, 1992) tell us that method study and work measurement are utilised to understand human work potential in terms of time spent on completing a task. (Taylor, 1911) showed and believed that through the application of these techniques, it is possible to quantify work. If the work can be measured, it can be managed. As time progressed so did technology as did the methods of work measurement and the tools required to fulfil those needs. I will discuss in this paper six of the ways that technology has advanced in the manufacture of contact lenses that were used to increase efficiency in my production environment.1.    Static Application upgrade for sustained yield stabilityContact lenses are manufactured in high volume by sandwiching a measured dose of monomer liquid between two halves of a plastic mould. These moulds are manufactured on two injection moulding machines (IMM). There are two halves a base curve (BC) and a front curve (FC).The FC is made from a Polystyrene based plastic and the BC is made from a Zeoner blended plastic. During the assembly of these parts a static charge needs to be applied to the BC to break the surface tension of the monomer to assist the adhesion of the monomer to the base curve. The resistance of the monomer to the adhesion generates voids in the lens which appear as centre hole defects (CH) at the automated lens inspection (ALI) further through the process.Traditionally the charge is applied via a charging bar and the metal pallet that carries the curves was grounded by passing over an aluminium brush to complete the circuit(fig.1). Over time these brushes wear due to constant contact or debris would build up leading to inefficient grounding and inadequate static application. Figure 1 Grounding brush under palletThis would lead to an increase in the CH losses at ALI. Figure 2 below is weekly CH average over the past 42 weeks of manufacturing Line 50, still operating on the static charge and brush system. Between weeks 3 and 10 there was issues with monomer leaking from the overflow drain causing monomer residue to adhere to underside of pallets and eventually build up around the static brush causing elevated levels of CH failures at ALI. Brushes were replaced every two weeks until the monomer issue was resolved. ALI stabilized for several weeks with CH hovering between 2% and 3% which is an acceptable level.  CH started to slowly creep up, the brushes were found to be worn and were replaced. The graph below shows the mean for the year to be 4%.Figure 2 Data gathered via Yield analysis system used in J&J Visioncare(Duclos, 1995) iterates that many manufacturing organizations are process-oriented and through continuous improvement we should never be satisfied with the process and should always look to improve. (Ohno, 1988) determined there were seven types of waste and one of which is the manufacture of defective products. To reduce the variation in CH a more robust static application system is required. According to (Shingo, 1986), mistakes that lead to defects can be engineered out of the processes and defects can be eliminated this way.A new production line, Line 53 was installed with an upgraded static system that was not reliant on a single static bar and brushes to complete the circuit. The static is applied directly to each curve via a spider system and the brushes replaced with steel spring brakes that are grounded to the frame to complete the circuit see figure 3. This eliminated the requirement to replace worn brushes and therefore eliminate the downtime required to replace them.Figure 3 Upgraded static application system with grounding brakes indicatedAs can be seen in figure 4 the levels of CH on line 53 with the newer static system averages out at 2.8% mean for the year to date and that includes an elevated level due to external issues with tooling on the IMM. The levels of CH rarely rise above 3% and when they do they are not static related and can be addressed accordingly. It is possible to roll this static system back to line 50 with some hardware changes and some minor software changes. Currently there is no plan to do so.Figure 4 Data gathered via Yield analysis system used in J&J Visioncare2.    Data Analytics/Mining/VisualizationUsing data analysis and visualization, companies can “manage and allocate their resources in a more effective and efficient way, reducing costs, and improving the quality of products and services that they offer (Solarte, 2002).” Automated manufacturing lines that are PLC controlled gather an enormous amount of data regarding the condition of the machine at any given time. This information is stored on servers and can be accessed through various interfacing software packages such as Tableau or Excel by relevant software engineers to give a graphical visualization of this data in a manner that can be utilized by users for decision making purposes.In figure 5 below is a screen shot of a Tableau system currently in use to analyse the data generated while manufacturing contact lens. The highlighted reject shows that it became an issue during the third hour 4 shifts prior, this gives an indication that there was something happening on the line at that time that caused that reject to increase. This information is vital as the technician can go back through the log and see what the driver for that issue was. Figure 6 shows some of the various visualizations that other departments utilize to assist in their decision making daily.Figure 5 Top N detractor Tableau Line 48Figure 6 some of the visualizations in use at J&J VisioncareFigure 7 (Solarte, 2002)3.    Array placement upgrade to reduce lossesDuring placement of an array, a term used to describe five blisters containing lenses floating in solution and sealed, there is a high risk of a product jam as the placement is reliant on gravity as arrays fall through an alignment chute, known as a waterfall placement. If there is an issue with placement and an array gets jammed in the waterfall chute this generates a product jam, which is referenced as a “zone 18 product jam” on the alarm data log. The difficulty in clearing the jam is the restricted space between the waterfall and the top of the tray, as can be seen in figure 9Figure 8 Waterfall style array placementData shows that in 2017 production line 47 lost almost 35 million lenses to product jams alone. This data was gathered from the line alarm log. See table 1. The line was taken out of production at the end of quarter 2 for the last two weeks of May and the month of June for product conversion and retro fitted with a new array placement system. This system included a flip station (fig 9) to present arrays to a second robot for direct placement into the trays (fig 10).  Because of this upgrade and the removal of the waterfall system, the lens per shift loss (LPS Loss) has been reduced from 7.8 million in quarter 1 to 2 million for quarter 3. With a manufacturing lens cost of $0.14(taken from JJVC CEP 2018 for Triton)that is a $1.1M loss of lenses in quarter 1 compared to a $290k loss in quarter 3.PartZ18_Product JamRow LabelsSum of Downtime MinsSum of LPS LOSS!20171204743481698620184095911837151Qtr1272077862823Jan80282320092Feb100982918322Mar90812624409Qtr254001560600Apr43201248480May1080312120Qtr371642070396Jul2790806310Aug2403694467Sep1971569619Qtr41188343332Grand Total16143346654137Table 1 Zone 18 product jams 2017/2018Figure 9 Flip station presenting arraysFigure 10 direct placement into trays4.    Acuity system utilised for array detectionTraditionally the placement confirmation of the waterfall array placement was completed by photo sensor which detected the array as it broke the beam. See fig 11Figure 11 Array presence sensorsThe main problem with these sensors is that they are powered by an amp and the feedback intensity is what determines an array present, thus can be adjusted and if the limits on the amp are not set correctly a false reading may lead to a miss count of arrays present and lead to an underage of product further down the line in the packing section. (Shingo, 1986)believed that defects could simply be eliminated from the outset instead of relying on measures taken on after-the-fact. So to eliminate the variable we needed to remove the reliance on the programmable amp and photo sensor.During the installation of the direct placement robot assembly the opportunity presented itself to upgrade the placement detection with a camera-based system. The system was already in use for the laser etching inspection of the lid stock. An expansion to the set-up was not a major overhaul.Three cameras were added to the package, one inspecting the tray for damaged ribs, damaged trays can cause issues in the placement of the arrays, and one for each tray. These were added to a multi-screen display along with the cameras inspecting the laser etched lid stock, see Fig 12Figure 12 Foil vision and array placement systemThis removed the variable that may have a quality implication further down the line and still be autonomous. (de Brouwer, 2011)describes autonomation as a “form of intelligent automation that has human characteristics. It is focused on detecting (potential) errors early in the process and signalling them such that they can be resolved by human intervention before they end up in the product.5.    Augmented reality as a training toolTraining technicians to acquire new maintenance and assembly skills is important for various industries. Because maintenance and assembly tasks can be very complex, training technicians to efficiently perform new skills is challenging. Training of this type can be supported by Augmented Reality, a powerful industrial training technology that directly links instructions on how to perform the service tasks to the machine parts that require processing. (Sabine Webel, 2012)Currently there is a project in progress to introduce augmented reality test and learning for process technicians. This will give them the ability to access context sensitive information/data at the right time to successfully complex tasks.TriggersChallengesConsequencesManuals in pdf formatIncreased duration of taskLower efficiency in task completionTechnician rotationLack of technician experienceIncreased number of errorsShorter training timeTechnician uncertain of taskProlonged equipment downtimeExpertise geographical dispersionIncreased duration of taskLower efficiency in task completionIncreased costsTable 2 challenges, consequences and root cause of increased costs in for tasks. (Amo, et al., 2018)One of the key factors in the training of technicians is the presentation of supporting data, installation/operating manuals etc. These are typically supplied in a pdf format which may contain a massive amount of data that needs to be dredged through to find the task applicable to the trainee. Often the technician may not be familiar with the information supplied in the manual and be reliant on knowledge from an experienced co-worker. (Amo, et al., 2018)in their study determined that animation was the preferred format of data that would be of most benefit in the transfer of knowledge. This animation would be mapped over real time video capture on a hand-held device such as a tablet/iPad(fig13)Figure 13 image courtesy of ShutterstockThe project currently in development is targeting the direct placement array transfer system. This system comprises of two robots a rotate station and a stock station area. (fig 14)This is a new type of system and the subject matter expertise is with a small number of technicians and training generally does not occur until there is a catastrophic failure of a robot and hard downtime is experienced while the robot is replaced. The training then becomes on-the-job training with added pressure of minimising downtime, best practice may be side-lined while the experienced technician uses their knowledge to speed up the installation and setup procedure.The AR training will cover the initial set-up of the robot pick and place positions, general routine maintenance tasks and commo error/fault rectification as per procedures. This work aid will take away the reliance on the large block of text in the operations manual and remove any ambiguity or misinterpretation of said data, in this case the complete teaching of the robot setup is 30 pages long.This is work and method study as defined by (Taylor, 1911) (Groover, 2014) and (International Labor Organisation, 1992) in its purest form, while also training a technician to a competent level. Taking the best-known method and applying the training through technology so the procedure remains constant.Figure 14 Direct placement system, Image taken from operations manual6.    Tween pad upgrade for increased yield and throughputContact lenses are manufactured in high volume by sandwiching a measured dose of monomer liquid between two halves of a plastic mould. These moulds are manufactured on two injection moulding machines (IMM). There are two halves a base curve (BC) and a front curve (FC). The measured monomer is injected into the front curve and the base curve is assembled onto the front curve and pressure applied. Any excess monomer is squeezed out and forms a ring of monomer known as a ‘hema ring’ that cures hard during the curing process (fig 15)Figure 15To ensure the hema ring is removed with the BC an inhibitor 9s applied to the FC. This inhibitor is a chemical known as Tween Span 80 and is light yellow viscose oily liquid. Span 80 is insoluble in water and soluble in organic solvents[1]. It is water/oil type emulsifier. it contains the monomer and stops it adhering to the FC. The application is made by means of a stamp and is applied to the FC before monomer dose. This is audited periodically by removing pallets before the dose and checking for the presence by applying a light powder, the powder adheres to the liquid(fig 16)Figure 16 Tween stamp auditWhen the applicators begin to wear, the efficiency of the application drops and there is an increase in hema ring rejects. To mitigate this an audit is carried out and downtime is experienced replacing the pads. A project was launched to reduce the level of hema rings and prolong the life of the pad. As this is a quality critical part of the process, the material of the stamping pad had to remain or there would be a lengthy change request process for the quality review board.The manufacturer was approached with specifications form three pads with different diameter applicators but retain the material specifications.Figure 17 excerpt from project test plan JJVC 2017The RD-03485.7 pad was installed on production line 42 to trial, results below show the step change and the step back up in hema when change reverted.Figure 18 trial line 42This style pad was rolled across two other lines manufacturing the same product to realise gains as illustrated in table 3LineShift Gain of LensesDaily GainRoll Up Annualized 360 Days426550130004,680,000505270105403,794,400538,900178006,408,000Total14,882,400Table 3ConclusionFigure 19 A 4-Stage IoT solutions architecture© HPE. (HughBoyes, 2018)Industrial engineers explore ways to manufacture products or complete tasks in a more efficient manner using various techniques. The most important tool they require is data, the more accurate the better. Even as technology advances, method study and work measurement remain a huge part of how continuous improvement is realised. Information, regardless of how it is gathered is the root of all improvement.  Technological advancement allows more and more data to be gathered and stored in servers waiting to be mined by data scientists so professionals in their various fields can make decisions based on the patterns that are uncovered. All through this paper the information presented regarding the field I work in was accessed through local servers and projects ongoing that I have been involved in with  process engineers/manufacturing engineers who have used the continuous improvement ethos to deliver more efficient methods through technological advances.ReferencesAmo, I. F. d., Erkoyuncu, J. A., Roy, R. & Wilding, S., 2018. Augmented Reality in Maintenance: An information-centred design framework. Procedia Manufacturing, Volume 19, pp. 148-155.de Brouwer, D. H. a. S., 2011. A LEAN Approach toward Automated Analysis and Data Processing of Polymers using Proton NMR Spectroscopy. SLAS TECHNOLOGY: Translating Life Sciences Innovation, 16(1), pp. 1-16.Duclos, S. a. L., 1995. JIT in services: a review of current practices and future directions for research. International Journal of Service Industry Management,, 6(5), pp. 36-52.Groover, M. P., 2014. Work Systems: The Methods, Measurement and Management of Work. New International ed. Harlow: Pearson.HughBoyes, B. H., 2018. The industrial internet of things (IIoT): An analysis framework. Computers in Industry, 101(October), pp. 1-12.International Labor Organisation, 1992. Introduction to work study. Fourth ed. Geneva: International Labor Office.Ohno, T., 1988. Toyota Production System – Beyond Large-Scale Production. New York: Productivity press.Sabine Webel, U. B. E. G. O. P., 2012. An augmented reality training platform for assembly and maintenance skills. Robotics and Autonomous Systems, 61(4), pp. 398-403.Shingo, S., 1986. Zero Quality Control: Source Inspection and the PokaYoke. Portland: Productivity Press.Solarte, J., 2002. A Proposed Data Mining Methodology and its Application to Industrial Engineering. [Online] Available at: http://trace.tennessee.edu/cgi/viewcontent.cgi?article=3549&context=utk_gradthes[Accessed 29 10 2018].Taylor, F. W., 1911. The Principles of Scientific Management. New York: Harper Bros.BibliographyAmo, I. F. d., Erkoyuncu, J. A., Roy, R. & Wilding, S., 2018. Augmented Reality in Maintenance: An information-centred design framework. Procedia Manufacturing, Volume 19, pp. 148-155.de Brouwer, D. H. a. S., 2011. A LEAN Approach toward Automated Analysis and Data Processing of Polymers using Proton NMR Spectroscopy. SLAS TECHNOLOGY: Translating Life Sciences Innovation, 16(1), pp. 1-16.Duclos, S. a. L., 1995. JIT in services: a review of current practices and future directions for research. International Journal of Service Industry Management,, 6(5), pp. 36-52.Ford, H., 2007. My Life and Work. Reprint ed. New York: Cosimo.Groover, M. P., 2014. Work Systems: The Methods, Measurement and Management of Work. New International ed. Harlow: Pearson.HughBoyes, B. H., 2018. The industrial internet of things (IIoT): An analysis framework. Computers in Industry, 101(October), pp. 1-12.International Labor Organisation, 1992. Introduction to work study. Fourth ed. Geneva: International Labor Office.Ohno, T., 1988. Toyota Production System – Beyond Large-Scale Production. New York: Productivity press.Perkins, J. S., 1997. The Quest, the newsletter of the Gilbreth Network. [Online] Available at: http://gilbrethnetwork.tripod.com/qv1n2.html[Accessed 20 February 2018].Sabine Webel, U. B. E. G. O. P., 2012. An augmented reality training platform for assembly and maintenance skills. Robotics and Autonomous Systems, 61(4), pp. 398-403.Shingo, S., 1986. Zero Quality Control: Source Inspection and the PokaYoke. Portland: Productivity Press.Solarte, J., 2002. A Proposed Data Mining Methodology and its Application to Industrial Engineering. [Online] Available at: http://trace.tennessee.edu/cgi/viewcontent.cgi?article=3549&context=utk_gradthes[Accessed 29 10 2018].Taylor, F. W., 1911. The Principles of Scientific Management. New York: Harper Bros.Toyota, 2018. The Origin of the Toyota Production System. [Online] Available at: https://www.toyota-global.com/company/vision_philosophy/toyota_production_system/origin_of_the_toyota_production_system.html