Maintenance Optimisation

All asset management systems are dynamic. Over time, they are subject to gradual decline in performance and will trend towards disorder. To fix this, you will have to optimize your approach and strategy. To optimize maintenance costs and improve plant availability through a well-balanced maintenance plan.

Using analytics for optimizing maintenance activity following questions can be answered

  • What are the bad actors?
  • What is the time to replace an equipment / part?
  • Number of time the failure has occurred and the time when it has occurred
  • Which components need to be replaced to keep my substation healthy?
  • The cost of an unplanned failure for a component, subject to a wear out failure mode, is twenty times the cost of a planned replacement.

Here are three major types of philosophies to consider.

  • Reactive Maintenance (Run-to-fail): You only take action to fix a piece after it breaks.
  • Preventive Maintenance (Time-based): You inspect equipment regularly to identify repairs and needed servicing before breakdowns occur.
  • Predictive Maintenance (Condition-based): This refers to evaluating the condition of components of equipment to decide whether it’s time for servicing.

The secret is to find the right balance of these which is the most cost-effective to best extend asset life.

Predictive maintenance is best used in conjunction with preventive maintenance and is most suited for use on your most critical, complex, and expensive equipment. An effectively executed maintenance plan will all but eliminate catastrophic failures. Independent studies show that breakdowns are reduced by 70% to 75% with a functional predictive maintenance plan.

Challenges faced in Maintenance activity

  • Production time loss due to unplanned downtime
  • Need for additional spare parts inventory
  • Damage to life or property be caused by uncontrolled change
  • Configuration and changes not documented or controlled
  • Need for service level standardization
  • Decentralized management

Traditional maintenance scheduling (reactive & planned) observes above challenges which costs the organization with time and money. Predictive maintenance scheduling (proactive) almost eliminates the above issues by continuous monitoring and ongoing maintenance, organizational changes and increase training which leads to longer lifespan of equipment, decreased downtime (planned or unplanned), more cost effectiveness, lower spare part inventory and many more due to which defour has adapted this technology.          

 

Defour’s Maintenance Optimization Model

Defour understands the need for optimizing maintenance activity and hence uses predictive modeling technique. Our model helps the organization to plan their maintenance effectively in a way that the equipment is worked to its full capability and there is minimum capital loss.

Our solution is basically reliability centered maintenance and quantitative trades are made between:

  • Scheduled and unscheduled maintenance,
  • Non-destructive inspections versus parts replacement,
  • Corrective action versus "do nothing,"
  • Different times-between-overhauls,
  • Optimal replacement intervals.

 

ILLUSTRATION OF DEFOUR’S MODEL

Below is the result of maintenance planning: -

Beta=0.9 indicates that the nature of the failure is around infant mortality to random hence current maintenance program may remain in vogue.

Beta=3.01 indicates we need very sound maintained program to avoid the failure.

Characteristic life is the expected life for the item with a high confidence interval.

Beta=0.9 indicates that the nature of the failure is around infant mortality to random hence current maintenance program may remain in vogue.

Beta=3.01 indicates we need very sound maintenance program to avoid the failure.

Characteristic life is the expected life for the item with a high confidence interval.

Beta, (β) tells the analyst whether or not scheduled inspections and overhauls are needed. For wear out failure modes, if the cost of an unplanned failure is much greater than the cost of a planned replacement, there is an optimum replacement interval for minimum cost.

On similar lines, our model will estimate the best close predictions for maintenance and also prescribe alternatives in case the plant has to be kept running to meet its production targets if any.  The model is build based on analytics and validated statistically.