Issue link: https://industrialengineer.epubxp.com/i/826827
48 ISE Magazine | www.iise.org/ISEmagazine Analytics for process design and improvement exactly what processes and resources would be needed at each milestone to keep pace with predicted demand. This deep insight means that the business had a robust framework for producing relevant predictive analytics that can pre-empt forecast changes in the business, ensuring that pro- ductivity remains high and that costs are minimized. It also means that it has a highly credible plan to present to internal and external stakeholders. Predictive simulation and analyt- ics has quickly become an invaluable tool in communicating with the organization's management team and company stake- holders. It can demonstrate the robustness of strategic plans to investors, providing transparency and instilling confidence to support more than $10 million of new capital investment. It has been invaluable in securing wider stakeholder engagement and buy-in, including the manufacturing workforce, subcon- tractors, customers and new sales prospects. It was a key dif- ferentiator in winning a recent multimillion-dollar contract. Success story 3: A worldwide automotive company used current asset performance analytics to understand the opera- tional and OEE implications of specific human interactions and decisions. This major production company has access to tremendous amounts of performance data across a wide range of production situations. To take advantage of this data and the statistical technology available, the organization assigned a de- sign team the challenge of modeling and simulating the impli- cations of past performance on future state production designs. This predictive simulation and analysis project specifically aimed to develop control logic for two production lines to optimize the decal processes applied to more than 2,000 parts across numerous models and vehicle options. This control logic needed to find a distribution logic that would optimize direct and indirect staffing levels for the two production lines. The team collected historical performance data across a wide range of production situations at extremely short performance intervals. Historical direct and indirect labor staffing was cor- related to this data, and statistical predictive models were de- veloped to identify periods of high labor performance. The analytics were used to explain all performance variations and then provide the specific data requirements for future state simulations. The team used these models to challenge the accepted staff- ing models. The new staffing models offered solutions with specific yet flexible labor and staffing assignments based on hourly volume, mix, changing labor content and seemingly unpredictable interferences. This integration into the orga- nization's traditional production monitoring systems ensured manageable solutions to a complicated staffing assignment. The intent of analytics is not to replace the current pro- duction monitoring system. Rather, analytics will ensure the system remains relevant as factors that influence performance change. The technical aspects of each of these remarkable im- provement programs have required a commitment from man- agement. The timing for these modeling improvement results has varied based on complexity, team commitment, informa- tion access and a variety of other project considerations. It has been shown that in general, significant team knowledge was gained in weeks, significant improvement gained in months and significant cost improvements were and will be sustained over the asset's remaining life. Simulation can outline your future Predictive simulation modeling has proven capable of deliver- ing a wide range of operational benefits, including improving and supporting significant production design and improve- ment initiatives. The effectiveness and tangible business im- pacts from advanced predictive analytics methods such as sim- ulation are enhanced by today's availability of better, relevant operational data and computing power. This means that such techniques are progressively becoming required tools to secure an organization's competitive advantage. This article highlighted several important considerations that should be addressed and suggested a systematic approach that will create a continuous pattern of improvement. Careful- ly adhering to the presented outline can provide a sustainable framework for an organization to develop analytical methods and business intelligence for production design and operation- al improvement. Timothy Sta sfield is preside t of IET I c., which has grow o 60 full-time employees a d has provided professio al i dustrial e gi eeri g support to more tha 00 orga izatio s a d 4,500 locatio s world- wide for 28 years. Sta sfield holds a B.S. i dustrial e gi eeri g, a MBA a d a Ph.D. i a ufacturi g ma ageme t a d is passio ate about operatio al a d process desig d improveme t. Ro da Massey is a professio al e gi eer a d vice preside t of IET I c. She has bee ith the compa y for 25 years. O e of her roles i volves usi g data a alytics to better u dersta d the past, prese t a d future states of orga izatio s. Because of her exte sive experie ce i imula- tio , computer programmi g a d database applicatio s, she has bee key resource for developi g clie t-tailored models to better u dersta d day-to-day operatio s a d to predict effects of cha ge, a d she has worked exte sively o he desig d developme t of predictive models i he ma ufacturi g e viro me t. A drew Aitke s chief operati g officer of La er Group a d presi- de t of La er I c., which provides predictive simulatio ech ology a d a alytical services to empower smarter busi ess operatio s. Aitke leads global sales a d delivery operatio s at the board level. Followi g early ma ageme t positio s with U ilever, Aitke oved i to ma - ageme t co sulta cy a d has held C-level positio s with i ter atio al ma ageme t co sulti g firms, worki g across a broad ra ge of geogra- phies, busi ess sectors a d customers that i clude Ford, Shell, Rolls- Royce, HP, America xpress, Cummi s a d Mars.