David Simchi-Levi spoke again, offering his view of planning and scheduling with an eye to operational excellence.
Operational excellence implies all the customary tools that we've become accustomed to, like Lean, Six Sigma, Toyota Production System, just in time, and so on. Toyota began its Lean journey in discrete manufacturing in the middle of the last century, and in Toyota's understanding, Lean implies just in time and autonomation, or "automation with a human touch." The automotive industry followed in the 70s (applied to comfort features), 80s (quality) and 90s (Lean, product variety and flexibility). Lean really gained widespread acclaim following publication of "The Machine That Changed the World." More recently, other industries, especially those with assembly operations (hi-tech, telecoms, even furniture) have followed in implementing Lean.
As a refresher, the key objectives of operations excellence focus around eliminating waste (focus on material, inventory, energy, defects, time), reducing cost (continuous improvements, coordination and synchronization between manufacturing execution with supply chain pains and customer orders) and empowering employees (workers can stop the line; fostering problem-solving capabilities). These objectives complement each other: low work in process implies immediate sensing of problems and empowering employees to solve problems and contribute to continuous improvement. And lean is about continuous improvement of either the production process or the supply chain -- it's not a one-time event.
The principles of Lean are taking a holistic view (focusing on the entire production process); attention to details (operational details matter strategically); control over WIP, e.g., through Kanban or CONWIP (where you control the entire inventory and WIP on the production line); reduce cycle times (e.g., eliminating long setups, reducing delays, coordinating machine maintenance activities, minimizing machine downtime); and production smoothing (the more we can smooth out production, e.g., through constant volume and constant product mix, the better and smoother the supply chain).
You can compare Lean to Mass Production: In MP, decisions are made hierarchically, versus decision making delegated to workers under Lean. In MP, WIP is a buffer to smooth production, versus inventory viewed as a source of waste that must be reduced in Lean. In MP, the environment is a constraint, while in Lean the focus is on removing constraints.
How does Lean support a smooth production schedule? Through flexibility (rather than buffer inventory as was the case in a mass production model), using integrated planning and scheduling, U-shaped production lines (which allow companies to take advantage of worker cross-training), and worker cross-training (versus the mass production model of each worker being trained on one specific task).
David addressed operational excellence in the process industry, which has a great number of built-in constraints that make it inherently difficult to apply Lean to production (e.g., high demand variability, complex manufacturing process, focus on performance management and cost control, and complex product mix changes). David talked about the use of the Plant PowerOps (PPO) solution to address these challenges. PPO offers a "smart" planning and scheduling system for the process industry at the plant/production-line level. The solution aims to give a company a "holistic" view of the entire production process, to assess detailed manufacturing models, optimize around business goals, coordinate planning with scheduling (as the planner and scheduler can use the same database, although they use different views/screens appropriate to their roles), and focus on optimization.
As an example, he cited a company in the fast-moving consumer goods (dairy) space, challenged by a high volume, high product mix; shared resources (production and cleaning equipment); volatile demand, with high service level agreements; regulatory requirements; cleaning in place; and requirements for traceability. The company compared their manual process to the optimized process in terms of inventory excess (where inventory exceeds maximum days of supply), and saw a clear improvement (to near zero) through the optimized process. The optimized process also realized a 2-5 percent improvement in operational time/net production time; an 8-25 percent reduction in cycle time (including a reduction in processing time due to a reduction in setups through better scheduling); and an average increase of 6 percent in throughput. Cleaning costs and changeovers saw a 10-40 percent reduction. This company used PPO for integrated planning and scheduling, starting in Mexico before moving out to Europe (Russia) and the U.S., France and then Mexico again, moving from one plant to another. The company itself cited benefits in four buckets: IT (allowed for full implementation/use of SAP APO), User Acceptance (broad user adoption), Manufacturing Targets and Organizational (collaboration between planners and schedulers).
David also covered issues around dynamic safety stock and process flexibility and their role in operational excellence. Dynamic safety stock focuses on safety stock within the production process. The standard approach of APS solutions is to compute safety stock based on target fill rates, demand variability, average production lead time and lead time variability, and then to generate a production plan using this safety stock. But what if the average production lead time is different from the planned production lead times? Either safety stock winds up being higher than necessary or does not cover the target fill rate. The solution is to combine optimizing lot sizes with optimizing safety stocks. In a case study David cited, safety stock was reduced (by about 25%) and stock levels were brought down (by about 10%).
In terms of process flexibility, the objectives are the same as for supply chain flexibility. David offered a hypothetical case study where flexibility was achieved through worker cross-training, or worker flexibility, as a way of enabling responses to changes in demand, variable processing times, and equipment breakdowns or other production disruptions. His example covered scenarios of no cross-training, full cross-training (where each worker is capable of doing any given job on the production line) and "2-skill cross training" (whereby each worker is capable of doing two jobs on the production line). The example suggested that "2-skill cross-training" can yield about 20% improvement in throughput from the production line, or, with "2-skill cross-training" plus limited buffer inventory, improvement in throughput plus about 85% improvement in buffer inventory.
|Supply Chain authority Andrew Reese is Editor of Supply & Demand Chain Executive. He has been invited by IBM PR to attend this show as a blogger and speaker. Like all other speakers, Andrew will receive all speaker benefits including travel and board.|