{"id":18679,"date":"2026-04-19T09:03:00","date_gmt":"2026-04-19T02:03:00","guid":{"rendered":"https:\/\/beup.space\/?p=18679"},"modified":"2026-04-22T11:19:55","modified_gmt":"2026-04-22T04:19:55","slug":"logistics-optimization-algorithm","status":"publish","type":"post","link":"https:\/\/beup.space\/en\/logistics-optimization-algorithm\/","title":{"rendered":"Evaluating Logistics Efficiency \u2014 A Strategy for Transitioning from Intuition to Algorithmic Optimization"},"content":{"rendered":"

\u23f1 11-minute read<\/p>\n

<\/p>\n

\n

Short answer:<\/strong> Algorithmic logistics optimization means replacing \"gut-feel\" decisions in production scheduling and transportation dispatch with two independent algorithm layers \u2014 Scheduling<\/strong> to sequence jobs by EDD\/SPT\/Johnson rules, and Routing<\/strong> to optimize vehicle routes via Nearest Neighbor + 2-opt with Time Window constraints. SMEs that apply this correctly can reduce mileage by 15\u201330%, eliminate 100% of time window violations, and save 100\u2013150M VND\/year for a 5-vehicle fleet \u2014 without investing in new infrastructure.<\/p>\n<\/div>\n

As an operations advisor, I keep seeing the same paradox at SMEs: we invest heavily in trucks and machinery, yet let those very assets depend entirely on the \"memory\" and \"experience\" of the dispatcher. The waste isn't just on the fuel bill \u2014 it hides in operational blind spots: late orders from wrong priorities, empty trucks from overlapping routes, overloaded vehicles from uncalculated total weights.<\/p>\n

This article provides an analytical framework and a 4-step roadmap for transitioning from intuition-based operations to algorithmic optimization \u2014 with concrete ROI case studies from Ph\u00fa Qu\u00fd Beverage (saving 144M VND\/year) and MedFast Pharma (preventing 355kg\/300kg overload). All techniques can be implemented first in Excel Solver, then upgraded to Python + OR-Tools when volume exceeds 20 delivery stops.<\/p>\n

\n \"S\u01a1
The logistics optimization mind map \u2014 6 pillars connecting intuition-based management to algorithmic systems.<\/figcaption><\/figure>\n
\n

Article contents<\/p>\n

    \n
  1. The Barrier: the hidden cost of \"gut-feel\" operations<\/a><\/li>\n
  2. Job Scheduling Optimization \u2014 5 priority rules + Johnson's Algorithm<\/a><\/li>\n
  3. Vehicle Routing Optimization \u2014 VRP, Nearest Neighbor, Time Window<\/a><\/li>\n
  4. Real ROI \u2014 case studies from Ph\u00fa Qu\u00fd and MedFast<\/a><\/li>\n
  5. Technology Roadmap \u2014 from Excel to OR-Tools<\/a><\/li>\n
  6. Strategic Recommendations for Operations Managers<\/a><\/li>\n
  7. Frequently asked questions<\/a><\/li>\n<\/ol>\n<\/div>\n
    \n

    The Barrier: the hidden cost of \"gut-feel\" operations<\/h2>\n

    <\/p>\n

    \n

    Scheduling vs Routing<\/strong> are two independent problems that are often conflated. Scheduling<\/strong> answers \"what to do first?\" \u2014 sequencing jobs on machines and staff. Routing<\/strong> answers \"which path to take?\" \u2014 optimizing vehicle routes. Confusing these two concepts is the root cause of waste in non-digitized SMEs.<\/p>\n<\/div>\n

    In non-digitized businesses, logistics management typically relies on personal experience \u2014 \"do what's convenient\", \"do what's urgent\". As order volume grows, the human brain cannot simultaneously process dozens of constraints like load capacity, delivery windows, production deadlines, and driver skills. Result: resource utilization rate drops \u2014 machines sit idle for lack of planning, trucks run empty for dozens of kilometers on overlapping routes.<\/p>\n

    \n \"Tr\u01b0\u1edbc
    Before: dispatcher overwhelmed, paper maps, yelling \"overloaded!\", \"we're late!\". After: optimized routes displayed on a dashboard with 20%+ better utilization.<\/figcaption><\/figure>\n

    Comparison: Gut-feel Management vs. Algorithmic Optimization<\/strong><\/p>\n

    \n\n\n\n\n\n\n\n\n\n
    Criterion<\/th>\nGut-feel Management<\/th>\nAlgorithmic Optimization<\/th>\n<\/tr>\n<\/thead>\n
    Core question<\/td>\n\"Do what's convenient\", \"do what's urgent\"<\/td>\n\"Which order minimizes delays?\", \"which route is shortest?\"<\/td>\n<\/tr>\n
    Resources<\/td>\nPersonal memory and habits<\/td>\nDigitize machines, staff, and vehicles as data<\/td>\n<\/tr>\n
    Unit of work<\/td>\nDiscrete tasks \/ individual orders<\/td>\nLinked production orders and delivery points<\/td>\n<\/tr>\n
    Constraints handled<\/td>\n1\u20132 simple constraints<\/td>\nMulti-constraints: load capacity, time window, deadline<\/td>\n<\/tr>\n
    Tools<\/td>\nNotebook, memory, manual Excel<\/td>\nExcel Solver, Python + OR-Tools<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n

    Verdict:<\/strong> Switching to algorithms isn't just changing tools \u2014 it's changing management thinking. Optimization must start with Scheduling<\/em> to free up warehouse\/factory capacity, before<\/em> executing Routing<\/em> in the field. Smart route planning for trucks means nothing if the goods aren't ready yet.<\/p>\n

    \n

    Job Scheduling Optimization \u2014 5 priority rules + Johnson's Algorithm<\/h2>\n

    Scheduling is not just queuing \u2014 it's a strategy to protect service commitments (SLA). To optimize a machine or production line, priority rules must be applied scientifically rather than by \"whoever's loudest\".<\/p>\n

    \n

    5 Priority Rules in Job Scheduling<\/p>\n