logistics-rules-to-optimization
Translate logistics and operations rules into optimization variables and constraints
Install / Use
npx skills add benchflow-ai/skillsbench --skill logistics-rules-to-optimizationInstalls into whichever agent you are using.
SKILL.md
Installable skill definition
Quality Score
Category
Development & EngineeringSupported Platforms
Tags
Our assessment of logistics-rules-to-optimization
logistics-rules-to-optimization scores 85/100 on our quality scale, 2542nd of 4,698 Development & Engineering skills we index.
Its SKILL.md is 11 KB long, well organised into 22 sections with 20 code examples: a thorough specification that gives an agent plenty to work with.
With 1,813 GitHub stars, it is one of the more widely adopted skills in the catalogue.
Maintenance, license and trust
- The repository was last updated about 4 months ago. That is recent enough to be usable, but agent tooling moves fast, so check the instructions against your agent's current version.
- Our last check on 2026-11-04 found the source still online.
- It is released under the Apache-2.0 license, a permissive license that allows use, modification and commercial use with attribution.
- Its trust signals score 98/100, with no cautions. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.
Safety scan
No issues foundOur scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful.
AI review by kimi-k2.7-code on 2026-10-01. Automated pattern scan on 2026-09-30. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.
logistics-rules-to-optimization compared with similar skills
All 4 of these similar skills score higher than logistics-rules-to-optimization; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| logistics-rules-to-optimization (this skill)by benchflow-ai | 85 | 1.8k | 4mo ago | SKILL.md |
| taste-skillby Leonxlnx | 100 | 116.8k | 8d ago | CLAUDE.md |
| ai-job-searchby ai-job-search-dev | 100 | 72.1k | today | CLAUDE.md |
| claude-howtoby luongnv89 | 100 | 47.3k | 2d ago | CLAUDE.md |
| lobehubby lobehub | 100 | 85.6k | today | MCP Server |
Frequently asked questions
- How do I install logistics-rules-to-optimization?
- Run
npx skills add benchflow-ai/skillsbench --skill logistics-rules-to-optimization. The install tabs above show the steps for each supported agent. - Which AI agents does logistics-rules-to-optimization work with?
- It is written for Universal, as a SKILL.md file. Other agents that read the same format can often use it too.
- Is logistics-rules-to-optimization safe to use?
- Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful. It is Apache-2.0-licensed and scores 98/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
- Is logistics-rules-to-optimization still maintained?
- The repository was last updated about 4 months ago. That is recent enough to be usable, but agent tooling moves fast, so check the instructions against your agent's current version.
Skill content
View source on GitHubname: logistics-rules-to-optimization description: Translate logistics and operations rules into optimization variables and constraints. Use when an operations problem describes vehicles, routes, depots, pickups, dropoffs, inventory, capacity, assignments, time windows, service targets, penalties, resource limits, or other business rules that need to become an optimization model.
Logistics Rules To Optimization
Use this skill when the problem statement gives operational rules in words and the agent must turn them into an optimization model.
The goal is not only routing. The same translation pattern applies to transportation, dispatch, rebalancing, warehouse moves, staffing, scheduling, assignment, capacity planning, production, and service-level problems.
Rule Translation Workflow
-
List the entities.
- Examples: vehicles, locations, depots, jobs, workers, machines, products, arcs, time periods.
-
Choose the decision state.
- Binary variables for yes/no choices.
- Integer variables for counts, loads, inventory, units moved.
- Continuous variables for time, flow, cost, utilization, or fractional quantities.
-
Convert each business rule into one of these patterns.
- Conservation: what enters equals what leaves, plus/minus changes.
- Capacity: quantity cannot exceed a limit.
- Linking: a quantity is allowed only if a binary decision is active.
- Assignment: exactly one, at most one, or at least one choice.
- Sequence: if one action follows another, update load/time/state.
- Compatibility: prohibit impossible combinations.
- Soft penalty: add slack for unmet demand or violation cost.
-
Add the objective last.
- Keep named components such as travel cost, labor cost, inventory penalty, unmet demand penalty.
-
Extract and independently validate the answer.
- Recompute routes, loads, assignments, inventory, penalties, and objective from the output data.
Variable Patterns
Selection and Assignment
Use binary variables when an option is selected.
x = {(i, j): model.addVar(vtype="B", name=f"x_{i}_{j}") for i in I for j in J}
Common rules:
# each item i assigned to exactly one option j
for i in I:
model.addCons(quicksum(x[i, j] for j in J) == 1)
# option j can handle at most capacity[j] items
for j in J:
model.addCons(quicksum(x[i, j] for i in I) <= capacity[j])
Route Arcs
Use binary arc variables when the order of visits matters.
x = {
(v, i, j): model.addVar(vtype="B", name=f"x_{v}_{i}_{j}")
for v in vehicles
for i, j in arcs
}
Use x[v, i, j] = 1 to mean vehicle/resource v goes directly from node i to node j.
Visit Indicator
Define visit from route arcs instead of creating a second binary unless the model needs it repeatedly.
visit = quicksum(x[v, i, j] for j in to_nodes if j != i)
If a standalone variable is useful:
visit = {(v, i): model.addVar(vtype="B", name=f"visit_{v}_{i}") for v in vehicles for i in locations}
for v in vehicles:
for i in locations:
model.addCons(visit[v, i] == quicksum(x[v, i, j] for j in to_nodes if j != i))
Quantity, Load, Inventory, and Time
load = {(v, i): model.addVar(vtype="I", lb=0, ub=vehicle_capacity, name=f"load_{v}_{i}") for v in vehicles for i in nodes}
service = {(v, i): model.addVar(vtype="I", lb=-vehicle_capacity, ub=vehicle_capacity, name=f"service_{v}_{i}") for v in vehicles for i in locations}
inventory = {(i, t): model.addVar(vtype="I", lb=0, ub=storage_capacity[i], name=f"inventory_{i}_{t}") for i in locations for t in periods}
arrival = {(v, i): model.addVar(vtype="C", lb=0, name=f"arrival_{v}_{i}") for v in vehicles for i in nodes}
Use integer variables for physical unit counts when the output must be integer-valued.
Common Logistics Rules
| Business Rule | Variable Choice | Constraint Pattern |
| --- | --- | --- |
| Choose exactly one option | x[i,j] binary | sum_j x[i,j] == 1 |
| Choose at most one option | x[i,j] binary | sum_j x[i,j] <= 1 |
| Open facility before assigning to it | open[j], assign[i,j] binary | assign[i,j] <= open[j] |
| Resource capacity | quantity variable | sum_i q[i,j] <= capacity[j] |
| Quantity only if selected | q[i], use[i] | q[i] <= M * use[i] |
| Fixed cost if used | use[i] binary | add fixed_cost[i] * use[i] to objective |
| Mutually exclusive modes | mode binaries | sum_m mode[i,m] <= 1 |
| Incompatible pair | two binaries | x[a] + x[b] <= 1 |
| Demand must be met | flow/quantity | supply_to[i] >= demand[i] |
| Demand may be unmet | nonnegative slack | served[i] + unmet[i] >= demand[i] |
| Absolute deviation penalty | nonnegative slack | actual-target <= dev, target-actual <= dev |
| Inventory balance | inventory variables | inv[t+1] = inv[t] + inbound - outbound |
| Station/storage upper bound | inventory variable | inv[i,t] <= capacity[i] |
| Cannot remove unavailable stock | move variable | outbound[i,t] <= inv[i,t] |
| Vehicle starts at depot | arc variables | sum_j x[v, START, j] == use_vehicle[v] |
| Vehicle ends at depot | arc variables | sum_i x[v, i, END] == use_vehicle[v] |
| Route continuity | arc variables | incoming[v,i] == outgoing[v,i] |
| Visit at most once | arc variables | outgoing[v,i] <= 1 |
| Split service allowed | arc/quantity variables | omit global single-visit; aggregate quantities over resources |
| Time window | arrival variable | earliest[i] <= arrival[v,i] <= latest[i] when visited |
| Travel time propagation | arc + arrival | arrival[j] >= arrival[i] + service_time[i] + travel[i,j] - M(1-x[i,j]) |
| Precedence | start/arrival variables | start[b] >= finish[a] |
| Route duration limit | arc variables | sum travel[i,j] * x[v,i,j] <= max_duration[v] |
Constraint Examples
Capacity
for r in resources:
model.addCons(quicksum(amount[i, r] for i in items) <= capacity[r])
Quantity Allowed Only When Active
Use the tightest possible M.
for i in items:
model.addCons(quantity[i] <= upper_bound[i] * use[i])
Soft Demand Satisfaction
unmet = {i: model.addVar(vtype="I", lb=0, name=f"unmet_{i}") for i in customers}
for i in customers:
model.addCons(served[i] + unmet[i] >= demand[i])
penalty_cost = quicksum(penalty[i] * unmet[i] for i in customers)
Absolute Target Deviation
Never use Python abs() on solver expressions.
dev = {i: model.addVar(vtype="C", lb=0, name=f"dev_{i}") for i in items}
for i in items:
model.addCons(actual[i] - target[i] <= dev[i])
model.addCons(target[i] - actual[i] <= dev[i])
Depot Start and End
If every vehicle must be used:
for v in vehicles:
model.addCons(quicksum(x[v, START, j] for j in locations) == 1)
model.addCons(quicksum(x[v, i, END] for i in locations) == 1)
If vehicles are optional:
use_vehicle = {v: model.addVar(vtype="B", name=f"use_vehicle_{v}") for v in vehicles}
for v in vehicles:
model.addCons(quicksum(x[v, START, j] for j in locations) == use_vehicle[v])
model.addCons(quicksum(x[v, i, END] for i in locations) == use_vehicle[v])
Route Continuity and At-Most-Once Visits
for v in vehicles:
for i in locations:
incoming = quicksum(x[v, j, i] for j in from_nodes if j != i)
outgoing = quicksum(x[v, i, j] for j in to_nodes if j != i)
model.addCons(incoming == outgoing)
model.addCons(outgoing <= 1)
This means vehicle v visits location i no more than once. It does not prevent a different vehicle from also visiting i.
Global Single-Visit Rule
Use only when the real rule forbids split service across vehicles/resources.
for i in locations:
model.addCons(
quicksum(x[v, i, j] for v in vehicles for j in to_nodes if j != i) <= 1
)
Do not add this rule when a large pickup/dropoff target may need multiple vehicles.
Load or State Transition Along Selected Arcs
If state[j] = state[i] + change[j] when arc (i, j) is used:
M = 2 * vehicle_capacity
for v in vehicles:
for i, j in arcs:
change_at_j = service[v, j] if isinstance(j, int) else 0
model.addCons(load[v, j] - load[v, i] - change_at_j <= M * (1 - x[v, i, j]))
model.addCons(load[v, j] - load[v, i] - change_at_j >= -M * (1 - x[v, i, j]))
This pattern works for load, arrival time, battery charge, inventory state, and other route-dependent state variables. Pick M from real variable bounds.
Time Windows
for v in vehicles:
for i in locations:
visit_i = quicksum(x[v, i, j] for j in to_nodes if j != i)
model.addCons(arrival[v, i] >= earliest[i] - horizon * (1 - visit_i))
model.addCons(arrival[v, i] <= latest[i] + horizon * (1 - visit_i))
for i, j in arcs:
if j in locations:
model.addCons(
arrival[v, j] >= arrival[v, i] + service_time.get(i, 0) + travel_time[i, j] - horizon * (1 - x[v, i, j])
)
Inventory Pickup/Dropoff Pattern
For rebalancing or material movement, define one signed service variable. Recommended convention:
service[v, i] > 0: pickup from locationi, vehicle load increases, location inventory decreases.service[v, i] < 0: dropoff to locationi, vehicle load decreases, location inventory increases.
service = {
(v, i): model.addVar(vtype="I", lb=-vehicle_capacity, ub=vehicle_capacity, name=f"service_{v}_{i}")
for v in vehicles
for i in locations
}
for v in vehicles:
for i in locations:
visit_i = quicksum(x[v, i, j] for j in to_nodes if j != i)
model.addCons(service[v, i] <= vehicle_capacity * visit_i)
model.addCons(service[v, i] >= -vehicle_capacity * visit_i)
for i in locations:
net_change = quicksum(service[v, i] for v in vehicles)
free_space = storage_capacity[i] - initial_inventory[i]
model.addCons(net_change <= initial_inventory[i]) # pickup cannot exceed stock
model.addCons(net_change >= -free_space) # dropoff cannot exceed space
If the target is a desired net pickup/dropoff:
unmet = {i: model.addVar(vtype="I", lb=0, name=f"unmet_{i}") for i in locations}
for i in locations:
net_change = quicksum(service[v, i] for v in vehicles)
model.addCons(net_change - target[i] <= unmet[i])
model.addCons(target[i] - net_change <= unmet[i])
Extract pickup/dropoff output as:
picked_up = max(service_value, 0)
dropped_off = max(-service_value, 0)
Objective Assembly
Build named components:
travel_cost = quicksum(distance[i, j] * x[v, i, j] for v in vehicles for i, j in arcs)
fixed_cost = quicksum(vehicle_fixed_cost[v] * use_vehicle[v] for v in vehicles)
penalty_cost = quicksum(penalty[i] * unmet[i] for i in customers)
model.setObjective(travel_cost + fixed_cost + penalty_cost, "minimize")
Related Skills
taste-skill
116.8kAgent skills that give coding agents taste: art direction, reference images and a browser check instead of generic AI pages
ai-job-search
72.1kThe job search that runs on your machine. AI job application framework built on Claude Code: evaluate postings, tailor CVs, write cover letters, prep interviews. Fork it and own it.
claude-howto
47.3kA visual, example-driven guide to Claude Code — from basic concepts to advanced agents, with copy-paste templates that bring immediate value.
lobehub
85.6k🤯 LobeHub - Super-Agents on Your Team. Add super-agents to the chats your team already uses.They take on work by themselves and get better every week.
Languages
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.
