An Intelligent Decision and Management Support Tool
1 like460 views
1. The document proposes an Intelligent Decision and Management Support Tool (IDMST) to help food production managers make quick decisions that balance costs and profits. 2. The IDMST would use real-time shopfloor data, simulations, and a knowledge base to evaluate decision options and recommend the most cost-effective actions. 3. Such a tool could help small- and medium-sized food companies improve productivity and decision-making without requiring complex and expensive systems.
An Intelligent Decision and Management Support Tool
- 1. An Intelligent Decision and Management Support Tool Dr Muna Hamdi School of Engineering and Design Brunel University
- 2. Process Flow Manual and Real-Time: Planning Scheduling Dispatch Resource Management (humans, machines and materials) including Line and Shift Management Forecasting, Planning and Scheduling Daily and weekly reports Commercial Management Promotions and New Product Development 6am Next day order- 9 am Business meeting Extra issues of concern such as missing ingredient would raise 24hours advance warning Purchasing Department Stores and Warehouses Stocks Human Resources Agencies An estimate of 20% yearly production growths 5 years strategy plan and 6 months line staff requirement forecast Quality Assessments and Control Technical Department Engineering (Maintenance & Technology development) Stores
- 3. Automatic Good Management ‘ Making the right decision’ Efficiency Productivity Utilisation Maximise profits Minimise cost Goals Customer satisfaction Enabling tools Useful Information (e.g. predicted and actual alarms & events, alternative schedules and due dates, quality report…) Data accuracy Resource Management (Labour, machine, material) Minimise down time Meeting due date Quality Factors to monitor and control on the shop floor Planning, scheduling taking the following into account: Minimise max completion time Minimise order waiting time (due date) Minimise number of waiting orders. Longest Best before end interval Reducing the number of MTS product to minimise storage cost. Minimise number of labour Balance Profit & Customer satisfaction? Real Time Minimise Wastages
- 4. Aim Improving the quality of decisions taken during food production process with the aim to reduce costs associated with production process and customer satisfaction. by providing techniques for real-time resource management product quality assessment traceability
- 5. What are we looking for? The key elements contributing to high labour and processing overheads within the food production industry.
- 6. Quality and Traceability Are important to both Consumer and Retailers. The end consumer and retailers are looking for the best quality at the same price or less! Hence a change in the decision making culture of the food production floor incorporating quality and traceability customisation as well as other KPFs, is required.
- 7. Key Factors Customer Product Operation The number & Type of customers Product type Promotions Seasonal Requirements Ingredients Staffing (Skilled) Maintenance Planning (Planned and Unplanned) Customer Relations (satisfaction & communication) Information Management Automation & Modernisation Resource Management Shelf Life Data quality & management Machine age & type
- 8. Identify key performance factors (KPF) Ranking KPFs according to their importance with respect to profit. Cost/Profit has been identified as the common base for analysis and comparison to assess system performance. KPFs will then be measured against this common denominator.
- 9. Decisions Management Improve Yields & Reduce Wastage Right-First-Time Quality Eliminate Coding Errors Real time performance Minimise Downtime Efficiency Productivity Utilisation Customisation Inventory Management Performance goals The degree of automation and modernisation , which includes Data quality and management. Machine age and type SW used Planning scheduling Prediction Resource management Man, machine & material Minimise costs Customer Satisfaction Traceability Influential factors Number of orders The number & Type of customers The type of product Day of the week, month and season Product promotions The variation in ingredients Staff availability, particularly skilled staff Maintenance (planned or sudden) Customer – Factory Interface . Maximise profits KPF
- 10. KPF Resource Utilisation deals with resource availability i.e. busy, idle, available, broken, shifts, and capacity issues . Customer Satisfaction will be measured as the function of variation between customer expectation and product/service provided . This includes packaging, content, traceability, freshness, and price . Productivity can be associated with Yield . Yield is defined as the variation in percentage of ingredients due to machine setups such as over-filling or seasonal variances such as oil absorption levels in cooking process. The variances in yield will not result into wastage and product rejection . Inventory Management includes levels of stock and spare parts . Efficiency deals with wastages e.g. rejected parts and energy.
- 11. What do we want to achieve? We are proposing a robust Intelligent Decision and Management Support Tool (IDMST) for food production managers to take quick-response measures in conflicting situations on daily basis. The tool proposes a cost-benefit analysis with regard to any action taken .
- 12. There is no quick response decision making tool in the industry to help production managers evaluate the gains and losses behind each option available to them. The food industry lacks the means of evaluating their decisions criteria in terms of costs and profit. Evaluate and predict future events and plans (Real-time shopfloor level).
- 13. The transformation process must be gradual taking into account the existing technologies To avoid any prohibitive costs and consequences .
- 14. Although our analysis addresses the global view of food manufacturing environment, this work focuses mainly on the daily management issues within the shopfloor activities .
- 15. How much would it cost? SME companies are not able to afford to invest on sophisticated SCADA and ERP systems nor do they need such general and complex systems. An easy to use cost effective solution could be more attractive to them.
- 16. The proposed system, should be able to selectively and intelligently process critical data from the shopfloor and other information The data are then refined and used to update the simulation and scheduling processes, to accurately predict near-term performance
- 17. Mobile Data Acquisition The mobile data sensors in most cases will mostly replace the manual paper work without altering the work flow routine.
- 18. Store- raw material Preparation Cooking Filling Distribution Storage Check availability and monitor freshness Quality check If below threshold then ALARM This include taste, freshness, Best before end requirement Indicate cleaning and prep. Time needed. Asses Attribute and choose a recipe Oil Absorption, cooking time, and pressure Quality check If below threshold then ALARM overcooked or undercooked, Quality check If below threshold then ALARM This includes proper packaging, cleanness & correct coding Quality check Control best before end according to ambient temperature Indicate cleaning and prep. Time needed. Quality check Transportation condition and speed Food processing quality oriented check
- 19. What Decision to make? “ These are your real-time problems needing corrective actions (data processing),” “ And these are the corrective actions and their cost in term of profit at your disposal (diagnostics). From that which is already known about the schedule of orders, availability of materials, capability of filling lines, engineering resource, operator availability, your best and most cost effective action, would be ……………..” This would mean that your current order scheduling and proposal completion time would be changed as follows ………….” Alternative actions ranked according to their cost are …………”
- 20. Simulation-What if? Shopfloor Mobile Data Points Data Mining Inference Deduction Profit (KPI Ranking) Production operator Scheduler (profit, meeting due date) Report System future status Yes System future status Schedule Verify Orders State evaluation How much? Shopfloor System Current status Knowledge Base Proposed Decision support and production management tool
- 22. Diagnosis is interesting but can only be beneficial when acted upon .
- 23. Publications Ali Mousavi, Muna Hamdi and M. Sarhadi , Knowledge-Based Quick-Response Decision Making in the Food Processing Industry, Int. J. Industrial and Systems Engineering, Vol. 2, No. 1, 2007. Muna Hamdi, Ali Mousavi and M. Sarhadi, An Intelligent Decision & Management Support Tool in IMSIO III, pp. 197-207, 28-29 June 05, ISBN 0 903440 34 2. Muna Hamdi, Ali Mousavi and M. Sarhadi, Process flow and operations management within food-processing factories (e-Track), PlanSig 2004, University College Cork, Ireland.
Editor's Notes
- #3: Production managers are challenged by the difficult task of incorporating all related information in a decision-making process while managing resources and solving various problems that involve several departments and different issues.
- #5: intelligent and user friendly operations management system that predicts, detects, and reacts in real-time to sudden events; forecasts with a high degree of confidence current trends, future needs and changes.
- #8: ICT Information and computing technology
- #12: Proof-of-principle that the software requires minimal, non-specialist modification when new products, plant or layouts are introduced into the factory, can automatically generate the necessary database modification within an acceptable number of replicate production runs
- #17: Information presentation in a simplified manner hastens human understanding and response to solving problems. such as production changes, breakdowns, and resource status. Proof-of-principle that the software requires minimal, non-specialist modification when new products, plant or layouts are introduced into the factory, and can automatically generate the necessary database modification within an acceptable number of replicate production runs It is envisaged that such uniquely combined computer technology would effectively be able to instruct, not only: “ These are your real time problems needing corrective actions (data processing),” but also, “And these are the corrective actions at your disposal (diagnostics). From that which is already known about the schedule of orders, availability of materials, capability of filling lines, engineering resource, operator availability, knowledge and experience, your best, most cost effective action , would be ……………..” This would mean that your current order scheduling and proposal completion time would be changed as follows ………….”